FOREX - cabafx.comcabafx.com/trading-ebooks-collection/JAMES L. BICKFORD - Forex... · FOREX...

FOREXSHOCKWAVE

ANALYSIS

JAMES L. BICKFORD

New York Chicago San FranciscoLisbon London Madrid Mexico City

Milan New Delhi San Juan SeoulSingapore Sydney Toronto

http://dx.doi.org/10.1036/0071498141

Copyright © 2008 by The McGraw-Hill Companies. All rights reserved. Manufactured in theUnited States of America. Except as permitted under the United States Copyright Act of 1976, nopart of this publication may be reproduced or distributed in any form or by any means, or storedin a database or retrieval system, without the prior written permission of the publisher.

0-07-159608-9

The material in this eBook also appears in the print version of this title: 0-07-149814-1.

All trademarks are trademarks of their respective owners. Rather than put a trademark symbol afterevery occurrence of a trademarked name, we use names in an editorial fashion only, and to the benefit of the trademark owner, with no intention of infringement of the trademark. Where suchdesignations appear in this book, they have been printed with initial caps.

McGraw-Hill eBooks are available at special quantity discounts to use as premiums and sales promotions, or for use in corporate training programs. For more information, please contactGeorge Hoare, Special Sales, at [email protected] or (212) 904-4069.

TERMS OF USE

This is a copyrighted work and The McGraw-Hill Companies, Inc. (“McGraw-Hill”) and its licensors reserve all rights in and to the work. Use of this work is subject to these terms. Except aspermitted under the Copyright Act of 1976 and the right to store and retrieve one copy of the work,you may not decompile, disassemble, reverse engineer, reproduce, modify, create derivative worksbased upon, transmit, distribute, disseminate, sell, publish or sublicense the work or any part of itwithout McGraw-Hill’s prior consent. You may use the work for your own noncommercial andpersonal use; any other use of the work is strictly prohibited. Your right to use the work may beterminated if you fail to comply with these terms.

THE WORK IS PROVIDED “AS IS.” McGRAW-HILL AND ITS LICENSORS MAKE NOGUARANTEES OR WARRANTIES AS TO THE ACCURACY, ADEQUACY OR COMPLETENESS OF OR RESULTS TO BE OBTAINED FROM USING THE WORK,INCLUDING ANY INFORMATION THAT CAN BE ACCESSED THROUGH THE WORKVIA HYPERLINK OR OTHERWISE, AND EXPRESSLY DISCLAIM ANY WARRANTY,EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO IMPLIED WARRANTIES OFMERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. McGraw-Hill and itslicensors do not warrant or guarantee that the functions contained in the work will meet yourrequirements or that its operation will be uninterrupted or error free. Neither McGraw-Hill nor itslicensors shall be liable to you or anyone else for any inaccuracy, error or omission, regardless ofcause, in the work or for any damages resulting therefrom. McGraw-Hill has no responsibility forthe content of any information accessed through the work. Under no circumstances shall McGraw-Hill and/or its licensors be liable for any indirect, incidental, special, punitive, consequential or similar damages that result from the use of or inability to use the work, even ifany of them has been advised of the possibility of such damages. This limitation of liability shallapply to any claim or cause whatsoever whether such claim or cause arises in contract, tort or otherwise.

DOI: 10.1036/0071498141

http://dx.doi.org/10.1036/0071498141

We hope you enjoy thisMcGraw-Hill eBook! If

you’d like more information about this book,its author, or related books and websites,please click here.

Professional

Want to learn more?

http://dx.doi.org/10.1036/0071498141

Contents

Acknowledgment xiIntroduction xiii

Part 1 Fundamental Studies 1

Chapter 1 Fundamental versus Technical 3Overview 3Technical Analysis 4Fundamental Analysis 4

Chapter 2 Interest Rates 7Overview 7Forecasting Model 8Fundamental Data 8Visual Representation 8Regression Results 10Observations 10

Chapter 3 Balance of Trade 11Overview 11Forecasting Model 11Fundamental Data 12Visual Representation 12Observations 12Descriptive Statistics 14

iii

For more information about this title, click here

http://dx.doi.org/10.1036/0071498141

Regression Results 14Interpretation 15Enhancements 15Differential Analysis 15Conclusion 16

Chapter 4 Consumer Price Index 17Overview 17Forecasting Model 18CPI Data 18Individual Linear Regressions 18Regression Results and Interpretation 20Independent Variable 21Currency Data 21Cross-Correlation 22Enhancements 23

Chapter 5 Summary of Fundamental Studies 25Overview 25Composite Modeling 25Euro Currency and Fundamentals 26

Part 2 Exchange Rate Determinants 29

Chapter 6 Electronic Broking 31Overview 31EBS 31Reuters 32Market Makers 33Order Flow 33Live Book 34

Chapter 7 Scheduled News Releases 37News Sources 37Domestic Economic Calendar 38Foreign Economic Calendar 38Usage 39

iv Contents

Part 3 Activity and Range 41

Chapter 8 Activity 43Overview 43Definition 44Equiactivity Chart 44Comparative Activity 45Single-Currency Activity 48Direction 48Observation 50

Chapter 9 Composite Activity Charts 51Time of Day 51Day of Week 52

Chapter 10 Range 55Absolute Range 55Relative Range 56Observations 58

Chapter 11 Composite Range Charts 59Time of Day 59Day of Week 61Average Interval Range 61

Chapter 12 Activity versus Range 63Forecasting Model 63Scatter Diagrams 63Regression Results 66Usage 66

Chapter 13 Inertia 67Overview 67Raw Currency Data 67Forex Inertia 68Inertia Charts 70Observations 72Momentum versus Inertia 72

Contents v

Part 4 Forex Wave Theory 75

Chapter 14 Review of Wave Principles 77Overview 77

Reversal Charts 78

Elliott Waves 79

Recurring Wave Patterns 80

Waves within Waves 81

Rules and Guidelines 82

Chapter 15 Cycle Preliminaries 85Definitions 85

Minimum Fluctuation Unit 86

Minimum Reversal Amount 87

Swing Reversal Algorithm 87

Chapter 16 Basic Three-Wave Cycles 91Overview 91

Impulse Cycle (ID � A) 92

Rectangle (ID � B) 92

Contracting Descending Triangle (ID � C) 93

Contracting Ascending Triangle (ID � D) 93

Contracting Symmetrical Triangle (ID � E) 93

Expanding Ascending Triangle (ID � F) 94

Expanding Descending Triangle (ID � G) 94

Expanding Symmetrical Triangle (ID � H) 95

Connector (ID � I) 95

Bear Cycle Conversions 96

Frequencies 96

Chapter 17 Forecasting the Third Wave 99Overview 99

Forecasting Wave Relationships 99

Practical Example 100

Caveat 101

vi Contents

Chapter 18 Multiple Wave Cycles 103Overview 103Overlapping Waves 103Naming Convention 105

Chapter 19 Forecasting the Fourth Wave 107Overview 107Frequencies 107Fourth-Wave Forecast 107

Chapter 20 Forecasting the Fifth Wave 111Pruning 111Frequencies 112Fifth-Wave Forecast 113Extended Impulse Cycle 113Potential Head and Shoulders 114Extended Contracting Triangle 115Extended Expanding Triangle 116

Chapter 21 Forecasting the Sixth Wave 119Overview 119Frequencies 119Extended Impulse Cycle 120Head and Shoulders 122Chimera Cycle 123

Chapter 22 Advanced Features 125Fractal Levels 125Pivot Charts 126Channel Lines 127Cycle Collapsing 128

Part 5 Shockwave Raw Data 131

Chapter 23 OHLC Bar Charts 133Overview 133Study No. 1: January 6, 2006 134Study No. 2: January 13, 2006 135

Contents vii

Study No. 3: January 20, 2006 136Study No. 4: January 27, 2006 137Study No. 5: February 3, 2006 138Study No. 6: February 10, 2006 139Study No. 7: February 17, 2006 140Study No. 8: February 24, 2006 141Study No. 9: March 3, 2006 142Study No. 10: March 10, 2006 143Study No. 11: March 17, 2006 144Study No. 12: March 24, 2006 145Study No. 13: March 31, 2006 146Study No. 14: April 7, 2006 147Study No. 15: April 14, 2006 148Summary of Statistics 149Range Ratio 150Observations 151

Chapter 24 Tick Charts 153Overview 153Study No. 1: January 6, 2006 154Study No. 2: January 13, 2006 155Study No. 3: January 20, 2006 156Study No. 4: January 27, 2006 157Study No. 5: February 3, 2006 158Study No. 6: February 10, 2006 159Study No. 7: February 17, 2006 160Study No. 8: February 24, 2006 161Study No. 9: March 3, 2006 162Study No. 10: March 10, 2006 163Study No. 11: March 17, 2006 164Study No. 12: March 24, 2006 165Study No. 13: March 31, 2006 166Study No. 14: April 7, 2006 167Study No. 15: April 14, 2006 168Summary of Statistics 169Observations 169

viii Contents

Part 6 Shockwave Swing Data 171

Chapter 25 Swing Charts 173Overview 173Study No. 1: January 6, 2006 174Study No. 2: January 13, 2006 175Study No. 3: January 20, 2006 176Study No. 4: January 27, 2006 177Study No. 5: February 3, 2006 178Study No. 6: February 10, 2006 179Study No. 7: February 17, 2006 180Study No. 8: February 24, 2006 181Study No. 9: March 3, 2006 182Study No. 10: March 10, 2006 183Study No. 11: March 17, 2006 184Study No. 12: March 24, 2006 185Study No. 13: March 31, 2006 186Study No. 14: April 7, 2006 187Study No. 15: April 14, 2006 188Summary of Statistics 189

Chapter 26 Shockwave Observations 191Overview 191Fractal Levels 192Maxima and Minima 193Swing Velocity 194Blind Trading 195Retracement Trading 197Informed Trading 197Caveat 198

Appendices 199A ISO Currency Pairs 201B Exchange Rates 207C Euro Currency 209D Global Banking Hours 211

Contents ix

E Monthly OHLC and Activity Charts 217F Daily OHLC and Activity Charts 229G Composite Activity Charts (Daily) 241H Composite Range Charts (Daily) 249I Composite Activity Charts (Weekly) 257J Composite Range Charts (Weekly) 269K Resources 281

Glossary of Forex Terms 283

Index 297

x Contents

Acknowledgment

I wish to thank Paul J. Szeligowski, friend and economic analyst, for hiseditorial assistance in the preparation of this book. His insightful rec-ommendations and novel ideas proved invaluable in researching thenature and occasionally cryptic relationships that arise when scrutinizingfinancial wave theories.

xi

Copyright © 2008 by The McGraw-Hill Companies. Click here for terms of use.

This page intentionally left blank

Introduction

Economically, a shockwave occurs in an otherwise well-behavedtime series when a violent breakout in price manifests itself in eitherdirection. Shockwaves can be either anticipated or unanticipated.Fortunately for spot currency traders, most of these breakouts are expected owing to scheduled news announcements released bygovernment agencies and some independent institutions.

Knowing when shockwaves will occur gives Forex traders atremendous analytical advantage in measuring, identifying, andcategorizing each shockwave’s unique properties. By scrutinizinga plethora of historical shockwaves, analysts can isolate and deter-mine recurring shockwave personality traits, which, in turn, maydirectly influence market entry timing and the size and positionof the trade.

ABOUT THIS BOOK

This book not only examines the shockwave itself (which occa-sionally lasts less than a minute or two after the news release) butalso investigates the reactive phase that immediately follows theshockwave. This is in a very broad sense the economic equivalentto Newton’s third law of motion: “For every action, there is anequal and opposite reaction.”

Numerous analytical tools required to unveil the mysteries ofForex shockwaves are also explained in this book. Some are old

xiii


techniques with new variations (such as adjustable-range studies),and others are relative newcomers to the arena of technical analy-sis (such as activity, inertia, composite charting, and Forex wavetheory).

Lastly, I should note that trading in the global foreign exchangemarkets recently has exceeded $2 trillion a day and is expected todouble over the next five years. Spot currency traders with a solidknowledge in a wide variety of analytical tools and shockwavebehavior will have an appreciable advantage over the novicesentering the world’s largest financial market for the first time. Thehighly visual approach presented in this book allows investors touse it as a computer-side reference guide while working online intheir currency trading platforms.

HOW THIS BOOK IS ORGANIZED

There are seven major divisions within this book.

PART 1: FUNDAMENTAL STUDIES

The subject of this book is shockwave analysis, which is a branchof technical analysis. Why, then, include a section on fundamen-tal analysis? Simple. Fundamental economic factors do influencecurrency prices, and even though they may not affect price fluc-tuations directly at the tick level, it is always expedient for techni-cal analysts to be aware of the forces that can adversely influencetechnical forecasts. Regression studies for interest rates, balanceof trade, and consumer price indexes are included as referencematerial.

PART 2: EXCHANGE RATE DETERMINANTS

This section examines price fluctuations at the very lowest possiblelevel, the electronic transaction level. Order flow not only affectsprice movements but also liquidity, volatility, range of trading, andtransaction costs. Economic calendars for the most significant news

xiv Introduction

releases are also presented in this section to assist speculators inscheduling their daily and hourly trading sessions.

PART 3: ACTIVITY AND RANGE

Composite charts and diagnoses based on activity, range, andtheir interaction are explored in this section. The study of bothactivity and range is crucial to uncovering the semicryptic forcesbehind spot currency shockwaves.

PART 4: FOREX WAVE THEORY

Wave analysis of economic cycles began in earnest during the 1930sand continues to evolve even today. An innovative addition to tech-nical analysis called Forex wave theory is reviewed in this section.Wave relationships, cycle nomenclature, and channel lines areexamined in detail.

PART 5: SHOCKWAVE RAW DATA

Traditionally, the most significant news releases have occurred at8:30 a.m. ET on Fridays. Fifteen practical studies from January 6,2006 through April 14, 2006 are presented as OHLC vertical barcharts and continuous tick charts. These are accompanied by corresponding statistical analysis for each chart.

PART 6: SHOCKWAVE SWING DATA

In this section, the swing reversal algorithm essential to reversalcharting is applied to the raw data presented in Part 5. This assiststraders in identifying recurring shockwave patterns and providesinsight into determining market entry and exit timing.

APPENDICES

The 11 appendices are an integral part of shockwave analysis andare intended to act as detailed reference tools while readers aretrading online.

Introduction xv

DISCLAIMER

I want to emphasize that spot and futures currency trading may notbe suited to everyone’s disposition. All traders must be keenly awareof the risks involved and of the consequences of poor trading habitsand/or mismanaged resources. Neither the publisher nor theauthor is liable for any losses incurred while trading currencies.

xvi Introduction

PART 1Fundamental

Studies


Chapter 1Fundamental

versus Technical

OVERVIEW

It is commonly accepted that there are two major schools ofthought when formulating a trading strategy for any market, be itsecurities, futures, or currencies. They are called fundamentalanalysis and technical analysis. The former is based on economic fac-tors, whereas the latter is concerned with price actions. Of course,most traders may opt to include elements of both disciplines whilehoning their personal trading strategy.

This book was written specifically for spot currency scalpers andday traders who favor technical analysis and are particularly inter-ested in wave theory forecasting and shockwave studies that areinvestigated in the bulk of this book. In Part 2 (Exchange RateDeterminants), I examine price fluctuations at the lowest conceiv-able arena, the order-transaction level.

However, I feel that any work on spot currency forecasting is notcomplete if the influence of fundamental factors is totally ignored.Hence this chapter has been included.

3


4 Part 1: Fundamental Studies

TECHNICAL ANALYSIS

The motivating belief among all technical analysts is that past andcurrent technical data (open, high, low, and closing prices) pro-vides sufficient information to forecast price fluctuations for theimmediate future. This genre of trader also believes that identifi-able price formations and patterns recur with such a high degreeof frequency as to warrant market entry orders. This microcosmicviewpoint, however, is based on a rather restricted set of input data:the open, high, low, and closing prices. Futures speculators alsohave the luxury of volume and open interest, two fields not avail-able to Forex traders owing to the decentralized nature of the for-eign exchange markets.

The small-cap short-term Forex trader who remains glued to thecomputer monitor for hours on end is essentially granted only onepiece of incoming information—the prevailing market price.However, there exists a myriad of techniques by which the collec-tion and storage of the streaming data can be massaged and coercedinto revealing critical properties and characteristics about theunderlying currency pair. It is this process that makes the employ-ment of technical analysis a potentially profitable occupation.

Forex shockwave analysis is an emerging branch within techni-cal analysis that relies heavily on the principles of financial wavetheory. Traders will find many groundbreaking innovations in thechapters that follow, several of which they may mold and modifyto meet their trading goals and ambitions.

FUNDAMENTAL ANALYSIS

Fundamental analysis is a study of the economy and is based onthe assumption that the supply and demand for currencies is aresult of economic processes that can be observed in practice andthat can be predicted. Fundamental analysis studies the relation-ship between the evolution of exchange rates and economic indi-cators, a relationship that it verifies and uses to make predictions.

For currencies, a fundamental trading strategy consists of strate-gic assessments in which a certain currency is traded based on vir-tually any criteria excluding the price action. These criteriainclude, but are not limited to, the economic condition of thecountry that the currency represents, monetary policy, and otherelements that are fundamental to economies.

The focus of fundamental analysis is on the economic, social, andpolitical forces that drive supply and demand. There is no single setof beliefs that guides fundamental analysis, yet most fundamentalanalysts look at various macroeconomic indicators such as economicgrowth rates, interest rates, inflation, and unemployment. Severaltheories prevail as to how currencies should be valued (see Part 2for details).

Fundamental versus Technical 5

Chapter 2Interest Rates

OVERVIEW

It is common knowledge that domestic interest rates have a directinfluence on price fluctuations of the corresponding underlyingcurrency pairs. The first statistical relationship to scrutinize isbased on the well-known maxim, “Smart money follows higherinterest rates.” Large corporations and institutions tend to con-vert liquid assets to the currency of the country with the higherinterest rate.

For example, assume that the prevailing U.S. prime rate is 5.0percent and that the equivalent British official rate is 4.75 percent.Both countries announce a new rate on the same day: The U.S.prime rate declines to 4.25 percent, whereas the British official rateadvances to 5.75 percent, a relative rally of 1.5 percent for theBritish rate.

Assume that the U.S. XYZ Corporation is a major importer/exporter for both American and British durable goods. After theadvance of the British rate, it now becomes more profitable forXYZ Corporation to deposit its USD cash on hand reserves inBritish banks, earning the higher interest rate. Therefore, XYZCorporation buys 10 million units of the GBPUSD currency pair,longing the pound and shorting the dollar.

7



FORECASTING MODEL

In my forecasting model, I will assume that the currency pair is thedependent variable and that some function of the two interest ratesis the independent variable. Specifically, I will assume that theinterest-rate relationship is additive rather than multiplicative:

GBPUSD � A(U.K. rate � U.S. rate) � B

where A and B are the ordinary least squares (OLS) linear regres-sion coefficients, slope and intercept, respectively. Note that in thismodel the independent variable is the difference between the twointerest rates. Also note that an advance in the U.K. rate will causean advance in the GBPUSD currency pair. Conversely, an advancein the U.S. rate will cause a decline in the same currency pair.

FUNDAMENTAL DATA

The U.S. interest-rate data were obtained from an independentfinancial portal called Money Café (www.moneycafe.com/library/prime.htm). The British interest-rate data were obtained from theBank of England (www.bankofengland.co.uk/statistics/index.htm).Monthly data for the time frame January 2000 through December2005 for the GBPUSD currency pair was obtained from DiskTrading, Ltd. (www.disktrading.com) on three compact disks.

VISUAL REPRESENTATION

In the upper portion of Figure 2-1, the U.S. prime rate is higherthan the Bank of England (BOE) official rate in all cases. At thebottom of the chart is the difference between the two rates. Thevertical scaling on the right represents interest rates expressed aspercentages.

The solid continuous line in Figure 2-2 represents the monthlyexchange rate for the GBPUSD currency pair from January 2000to December 2005. The dotted line is the regression estimate forthe forecasting model described earlier.

www.moneycafe.com/library/prime.htm

www.moneycafe.com/library/prime.htm

www.bankofengland.co.uk/statistics/index.htm

www.disktrading.com

Interest Rates 9

10.0

US Prime Rate versus BOE Official RateJan 2000 to Dec 2005

US Rate minus BOE Rate

9.59.08.58.07.57.06.56.05.55.04.54.03.53.02.52.01.51.00.50.0

5.004.754.504.254.003.753.503.253.002.752.502.252.001.751.50

JF J AS J ASFONDJMAMU J AS FONDJ J AS ONDJ J AS J ASONDJ ONDONDJMAMU F MAMU F MAMUMAMU FMAMU

Figure 2-1 U.S. versus Bank of England interest rates and difference.

1.9400

GBPUSD Monthly Average and Regression EstimateJan 2000 to Dec 2005

1.92001.90001.88001.86001.84001.82001.80001.78001.76001.74001.72001.70001.68001.66001.64001.62001.60001.58001.56001.54001.52001.50001.48001.46001.44001.42001.40001.38001.36001.34001.32001.30001.2800

JF J AS JASFONDJMAMJ JAS FONDJ JASONDJ JAS JASONDJ ONDONDJMAMJ F MAMJ F MAMJMAMJ FMAMJ

Figure 2-2 GBPUSD and regression line.

REGRESSION RESULTS

A linear regression for the preceding model generated the resultsshown in Table 2-1.

OBSERVATIONS

First, I note that a coefficient of correlation of 62.39 is relativelyencouraging, considering all the other economic factors that influ-ence the GBPUSD exchange rate. Essentially, this means that 62percent of the directional movement in the underlying currencypair can be attributed directly to the fluctuations between the inter-est rates of the two countries.

Sadly, though, this information does not directly benefit daytraders and scalpers who operate at the tick level. The U.S. primeinterest rate is published by the Federal Reserve Bank of St. Louisand is only modified every one and a half to two and a halfmonths (on average). Thus critical analysis of interest rates is usu-ally left to long-term traders.


Table 2-1 Interest-Rate Regression Coefficients

Regression Coefficient Value

Slope 0.1861Intercept 2.1597Coefficient of correlation 62.39

Chapter 3Balance of Trade

OVERVIEW

The purpose of this chapter is to quantify the influence of the balance of trade between two countries on their correspondingcurrency pair. Balance of trade (BOT) is defined as

BOT � exports � imports

The balance of trade hypothesis is as follows: An increase in onecountry’s trade deficit will cause a proportional decrease in thatcountry’s currency. Conversely, as a country’s exports increase inrelation to its imports, the value of its currency also increases. Justas a point of curiosity, the United States has a deficit trade balancewith over 90 percent of the world’s nations.

FORECASTING MODEL

The following linear model will be employed using the USDCADcurrency pair:

USDCAD � A � (U.S. exports � CA imports) � B

An ordinary least squares (OLS) linear regression will be usedto extract the following:

11



A � first regression coefficient in the preceding model (slope)

B � second regression coefficient in the preceding model(intercept)

s � standard deviation

r � coefficient of correlation

see � standard error of the estimate

The critical value here is r, the coefficient of correlation, whichexplains how closely the currency pair (the dependent variable)follows the BOT data (the independent variable).

FUNDAMENTAL DATA

The U.S.–Canada BOT data were obtained from the U.S. CensusBureau, Department of Foreign Trade Statistics (www.census.gov/foreign-trade/balance/index.html). Initially, monthly data for thetime frame January 2000 through December 2005 were used. TheUSDCAD currency pair data were obtained from Disk Trading,Ltd. (www.disktrading.com) on three compact disks.

VISUAL REPRESENTATION

The first step is to chart the raw data to determine if there are anystatistical anomalies that may bias the regression coefficients andthe coefficient of correlation (Figures 3-1 and 3-2).

OBSERVATIONS

In Figure 3-1, on first sight we note that the BOT values exhibit arather “choppy” behavior. On closer inspection, we see that thissupposed erratic property is actually the sum of multiple sinu-soidals overlaid on top of a basic downward trend. This becomesan excellent candidate for trigonometric dissection at a later date.

www.census.gov/foreign-trade/balance/index.html

www.census.gov/foreign-trade/balance/index.html

www.disktrading.com

Balance of Trade 13

-3.2200

US-Canada Monthly Balance of Trade (in USD Billions)Jan 2000 to Dec 2005

-3.4200-3.6200-3.8200-4.0200-4.2200-4.4200-4.6200-4.8200-5.0200-5.2200-5.4200-5.6200-5.8200-6.0200-6.2200-6.4200-6.6200-6.8200-7.0200-7.2200-7.4200-7.6200-7.8200-8.0200-8.2200

JF AS ASFONDJMAMJJ AS FONDJ AS ONDJ AS ASONDJ ONDONDJMAMJJ FMAMJJ FMAMJJMAMJJ FMAMJJ

Figure 3-1 U.S.–Canada balance of trade.

1.6000

USDCAD Currency Pair - Monthly MidrangeJan 2000 to Dec 2005

1.58001.56001.54001.52001.50001.48001.46001.44001.42001.40001.38001.36001.34001.32001.30001.28001.26001.24001.22001.20001.18001.1600

JF AS ASFONDJMAMJJ AS FONDJ ASONDJ AS ASONDJ ONDONDJMAMJJ FMAMJJ FMAMJJMAMJJ FMAMJJ

Figure 3-2 USDCAD monthly midrange.

In Figure 3-2, I explain the arbitrary use of the monthlymidrange instead of the monthly close for one important reason:Through experimentation, it improved the coefficient of correla-tion by 2 percent on average. Midrange is a central value, whereasthe closing price can fluctuate anywhere inside the monthly range.

I also must note that a clear linear trend is not required in eitherset of data because the two sets are regressed against each otherand not against a straight line.

DESCRIPTIVE STATISTICS

The descriptive statistics for the independent variable (BOT) arelisted in Table 3-1.

The descriptive statistics for the dependent variables (USD-CAD) are listed in Table 3-2.

Descriptive statistics are used by analysts to parameterize vari-ous properties of the underlying data.

REGRESSION RESULTS

The OLS linear model generated the results shown in Table 3-3.


Table 3-1 Statistics for Balance of Trade

Statistic Value

High �3.1234Low �8.2129Range 5.0895Midrange �5.6682Mean �4.8273Standard deviation 1.1114

Table 3-2 Statistics for USDCAD Currency Pair

Statistic Value

High 1.6008Low 1.1591Range 0.4417Midrange 1.3799Mean 1.4202Standard deviation 0.1379

Therefore, the forecasting model has the arithmetic form

USDCAD � 0.0872 � BOT � 1.8411

INTERPRETATION

The surprisingly high coefficient of correlation implies that theBOT between the two countries does have a direct and positive cor-relation with the corresponding currency pair on a long-term basis(or at least in the U.S.–Canada example).

However, I must note that BOT is only one of many influenceson currency prices. The ultimate equation will have several depen-dent variables on the right side and require a multiple (nonlinear)regression for its solution.

ENHANCEMENTS

This study is far from being conclusive because it represents a sin-gle fundamental influence and a single currency pair. The sameapproach must be applied to all the major currency pairs and theircross rates, as well as numerous other fundamental categories.

Additional analysis on USDCAD balance of trade can includethe following.

DIFFERENTIAL ANALYSIS

This involves performing a linear regression on the monthlychange in the two data sets using the linear differential model:

USDCADcurrent � USDCADprevious � A � (BOTcurrent

� BOTprevious) � B

Balance of Trade 15

Table 3-3 Model Coefficients

Regression Coefficient Value

Slope 0.0872Intercept 1.8411Standard error of the estimate 0.0082Coefficient of correlation 70.29

SERIAL CORRELATION

Serial correlation techniques are employed to determine if aleader/lagger relationship exists between the two data sets. Thisis also called a dominance/dependence relationship.

The dependent variable (the currency pair) is shifted one timeunit (one month, in this case) into the future. This is to determineif the current BOT has a robust forecasting influence on the fol-lowing month’s currency pair. For example,

USDCADJuly � A � (BOTJune) � B

In the serial model, the coefficient of correlation must be sig-nificantly higher than the coefficient of correlation using the stan-dard nonserial model.

SMOOTHING TECHNIQUES

It is also expedient to determine the effect on the coefficient ofcorrelation when either the currency data or the BOT data or bothare massaged with mild moving-average filters.

TRIANGULAR CORRELATIONS

This involves the USD and two other currencies plus BOT dataamong all three corresponding countries. The mathematics ismore complex but may yield a significant increase in the aggre-gate coefficients of correlation.

CONCLUSION

A coefficient of correlation of 70 percent is very significant.However, spot currency traders must keep in mind that this infor-mation is used primarily by medium- and long-term speculators.


Chapter 4Consumer Price

Index

OVERVIEW

The purpose of this study is to quantify the influence of the consumer price index (CPI) between two countries on their cor-responding currency pair.

Fundamentally, the CPI (also called the retail price index) is a sta-tistical measure of a weighted average of prices of a specified setof goods and services purchased by wage earners in urban areas.It is a price index that provides a measure of inflation and a cost-of-living index. The CPI can be used to track changes in prices ofall goods and services purchased for consumption by urban house-holds. User fees (such as water and sewer service) and sales andexcise taxes paid by the consumer are also included. Income taxesand investment items (such as stocks, bonds, life insurance, andhomes) are not included. There are numerous methods to quan-tify one country’s inflation rate. In fact, a robust inflation indica-tor should incorporate over half a dozen or so such indicators.However, for the sake of clarity and simplicity, I have decided tofocus on the single-most important inflation indicator—the CPI.

The testing premise is simple: As one nation’s CPI rises, its currency decreases proportionately against foreign currencies.

17



FORECASTING MODEL

In this study I compare monthly CPI data between the UnitedStates and the United Kingdom; thus the currency pair GBPUSDis required. The method will be an OLS linear regression (cross-correlation) of the following model:

GBPUSD � A � (U.S. CPI/U.K. CPI) � B

Note that because of the inverse relationship defined in the pre-ceding hypothesis, the U.K. CPI must be in the denominator ofthe independent variable.

CPI DATA

The U.S. CPI data were downloaded from the Bureau of LaborStatistics (U.S. Department of Labor) at www.bls.gov/cpi and spanthe time period January 1971 through December 2005. The corre-sponding U.K. CPI data were downloaded from the National StatisticsWeb site at www.statistics.gov.uk/statbase/TSDdownload2.asp andcover the period January 1988 through December 2005.

The U.S. CPI data use a basis month of July 1983, whereas theU.K. CPI data use May 2005 as the basis month. The basis month iswhen the agency who compiled the data sets the CPI to 100.0 as areference point. This is an arbitrary decision on the part of thecompiling agencies, who, for very valid reasons, readjust the indexperiodically. Mathematically, the use of two different basis monthsin two sets of data has no effect on their mutual coefficient of cor-relation because scaling is linear.

INDIVIDUAL LINEAR REGRESSIONS

Before performing a cross-correlation between the two CPI datasets (the independent variables) and the corresponding cur-rency pair (the dependent variable), it is expedient to test theCPI data individually using a simple time-based linear model (see Figures 4-1 and 4-2):

CPI � A � MonthNo � B

www.bls.gov/cpi

www.statistics.gov.uk/statbase/TSDdownload2.asp

Consumer Price Index 19

195.0

US Monthly Consumer Price Index (July 1983 = 100)Jan 1988 to Dec 2005

190.0185.0180.0175.0170.0165.0160.0155.0150.0145.0140.0135.0130.0125.0120.0115.0

Figure 4-1 U.S. consumer price index.

GB Monthly Consumer Price Index (May 2005 = 100)Jan 1988 to Dec 2005

101.098.096.093.591.088.586.083.581.078.576.073.571.068.566.063.561.0

Figure 4-2 U.K. consumer price index.

REGRESSION RESULTS AND INTERPRETATION

Tables 4-1 and 4-2 present the regression results.The regression results of the two CPIs taken individually are

hardly surprising.First, the rates of increase (slope) over the 216-month time frame

are 2.81 and 5.73 percent, respectively (although the U.S. rateseems slightly low compared with other U.S. inflation indicators).

Second, both CPIs exhibit a very high coefficient of correlation(99.7 and 96.3 percent, respectively), which seems very consistentwith the linear model expectations.

Third, ignore the negative values for the intercept. Those aresimply hypothetical values if the regression line were drawn all theway back to the year 1 A.D. (actually the year 0, if it existed).


Table 4-1 U.S. Consumer Price Index: Linear Regression Results

Regression Statistic Value

High 199.2000Low 115.7000Range 83.5000Midrange 157.4500Mean 157.7046Standard deviation 22.1465Slope 2.8146Intercept –335.3753Coefficient of correlation 99.7368

Table 4-2 U.K. Consumer Price Index: Linear Regression Results


High 101.0000Low 61.9000Range 39.1000Midrange 81.4500Mean 86.0630Standard deviation 10.5155Slope 5.7250Intercept –384.2116Coefficient of correlation 96.3252

INDEPENDENT VARIABLE

Figure 4-3 and Table 4-3 show the independent variable anddescriptive statistics, respectively.

CURRENCY DATA

Raw GBPUSD currency data were obtained from Disk Trading,Ltd., on compact disks (see Figure 4-4 and Table 4-4).


Table 4-3 Descriptive Statistics for the Independent Variable

Statistic Value


US CPI / GB CPIJan 1988 to Dec 2005

1.8251.8351.8451.8551.8651.8751.8851.8951.9051.9151.9251.9351.9451.9551.9651.975

1.8151.8051.7951.785

1.7651.775

1.7551.7451.735

Figure 4-3 Independent variable U.S. CPI/U.K. CPI.

CROSS-CORRELATION

I will now return to my original model for the relationship betweena currency pair and the two CPIs of the corresponding countries:

GBPUSD � A � (U.S. CPI)/(U.K. CPI) � B

The regression yielded the results shown in Table 4-5.


GBPUSD Currency Pair–Monthly MidrangeJan 1988 to Dec 2005

1.55001.57501.60001.62501.65001.67501.70001.72501.75001.77501.80001.82501.85001.87501.90001.92501.9500

1.52501.50001.47501.45001.42501.4000

Figure 4-4 GBPUSD currency pair.

Table 4-4 Descriptive Statistics for the GBPUSD Currency Pair

Statistic Value


A purely visual examination of the preceding four charts indi-cates that a direct linear correlation would not be high. The 23 per-cent coefficient of correlation might appear low, but it is noteworthyto keep in mind that CPI data are only one of a half dozen or soindicators in the composite inflation index.

ENHANCEMENTS

There are experimental techniques to test the model further,such as

1. Convert the model from a multiplicative to an additive paradigm:

GBPUSD � A � (U.S. CPI – U.K. CPI) � B

2. Test a variety of logarithmic relationships:

log(GBPUSD) � A � (U.S. CPI)/(U.K. CPI) � B

GBPUSD � A � log[(U.S. CPI)/(U.K. CPI)] � B

log(GBPUSD) � A � log[(U.S. CPI)/(U.K. CPI)] � B

3. Segment both currency and CPI time series into annual sam-ples, and apply the original model on each year.

4. Test for any serial correlations in the currency data and filterthem out.

5. Combine the CPI data with another inflation indicator beforeperforming the cross-correlation.


Table 4-5 Cross-Correlation Coefficients


High 1.9782Low 1.7354Range 0.2427Midrange 1.8568Mean 1.8297Standard deviation 0.0615Slope 0.4955Intercept 0.7329Coefficient of correlation 22.81

Inferential statistics is a wide-open field of investigation. “If youtorture the data long enough, it is bound to say something.”

In other words, it is not the magnitude of the coefficient of cor-relation that is tantamount. It is whether the same results can bereproduced consistently with different time frames and otherminor modifications.


Chapter 5Summary of

FundamentalStudies

OVERVIEW

The preceding chapters examined the coefficients of correlationfor the linear forecasting models shown in Table 5-1.

In all three studies, it should be noted that only one funda-mental economic influence was regressed per forecasting model.

COMPOSITE MODELING

Forecasting long-term currency prices using fundamental data isa complex and sometimes ineffective endeavor. A more realisticforecasting model requires more independent variables on theright side of the modeling equation. For example,

25

Table 5-1 Model Correlation Summary

Study Forecasting Model Correlation

Interest rates GBPUSD � A(U.K. rate � U.S. rate) � B 63 percentTrade balance USDCAD � A(U.S. exports � CA imports) � B 70 percentInflation rates GBPUSD � A(U.S. CPI)/(U.K. CPI) � B 23 percent



GBPUSD � A(INT) � B(INF) � C(BOT) � D(PPP)� E(GDP) � F(PPI) � . . . � Z

where

INT � interest rates

INF � inflation rates

BOT � balance of trade

PPP � purchasing power parity

GDP � gross domestic product

PPI � producer price index

. . . � other fundamental influences such as housing starts,unemployment rate, etc.

The letters A through F are called the partial regression coefficientsfor a multiple regression. The letter Z represents a single intercept con-stant that applies to all the individual economic factors in the model.

The composite model can be solved using a technique calledthe solution to simultaneous equations. This may require specializedsoftware that most standard statistical packages include. Specifi-cally, the program for the Gauss-Jordan elimination technique isrecommended.

Again, it should be noted that historical technical data are avail-able for the smallest possible time unit—streaming tick data andinterval data as small as 1 minute. Historical fundamental data, onthe other hand, usually are packaged as weekly or monthly dataand, in some cases, quarterly data. Thus, as mentioned earlier,knowledge of fundamental influences can be its own reward, butit does not directly assist day traders and scalpers who operate atthe tick level in their online currency platforms.

EURO CURRENCY AND FUNDAMENTALS

When devising fundamental forecasting models that involve theeuro currency, the analyst should be aware that as of January 1,2001, there are 12 nations (13 if the Vatican is included) that gaveup their legacy currencies and switched to the euro.

Essentially, analysts will need fundamental data from all 13nations to forecast a currency pair estimate accurately. For exam-ple, assume that the analyst is investigating how BOT affects theEURUSD currency pair. Export and import data must be gatheredfor all 13 euro countries as they relate to commercial trade withthe United States. This is somewhat of a Herculean task for anindependent small-cap investor.

Summary of Fundamental Studies 27

PART 2Exchange RateDeterminants


Chapter 6Electronic Broking

OVERVIEW

Before delving into Forex wave theory and shockwave analysis, Iwant to present a brief review of how exchange rates are handledat the lowest possible transaction level—the automated electronic order-matching system.

EBS

Electronic Broking Services (EBS) is the world’s leading providerof Forex and precious metals transactional trading and data solu-tions to thousands of trading professionals and organizations.EBS was created in September 1993 by a partnership of the world’slargest foreign exchange market-making banks. Approximately$145 billion in spot foreign exchange transactions, 700,000 ouncesof gold, and 7 million ounces of silver are traded every day overthe EBS Spot Dealing System.

In June 2006, EBS was acquired by Intercapital Public LimitedCompany (ICAP), the world’s largest interdealer broker. Thisacquisition combined EBS’ strengths in electronic spot foreignexchange with ICAP’s electronic broking business to create a sin-gle global multiproduct business. ICAP has a daily average trans-action volume in excess of $1 trillion, 50 percent of which is

31



electronic. ICAP provides specialist intermediary broking servicesto commercial banks and investment banks in the wholesale finan-cial markets.

An interdealer broker works in these markets to draw togetherliquidity and to match buyers and sellers so that deals can be exe-cuted by its banking customers. Banks pay a commission when theyuse a broker to complete a deal.

REUTERS

The Reuters Group is best known as a news service that providesreports from around the world to newspapers and broadcasters.However, news reporting accounts for less than 10 percent of thecompany’s income. Its main focus is on supplying the financialmarkets with information and trading products. These includemarket data, such as share prices and currency rates, research andanalytics, as well as trading systems that allow dealers to buy andsell currencies and shares on a computer screen instead of by tele-phone or on a trading floor such as that of the New York StockExchange. Reuters has offered a traditional electronic platformsince 1989.

Reuters and EBS compete head-on for the liquidity offered bymajor foreign exchange banks. Reuters is fighting back with thelaunch of its upgraded Dealing 3000 screen-based system. But bothEBS and Reuters give access to a relatively limited number of play-ers, focusing on the interbank market.

The Reuters Dealing 3000 system is changing over from the oldproprietary front end used on Dealing 2000 to a Windows NT envi-ronment with an improved look and flexible screen layout. Itsother new features include a broadcast facility, the ability to carryout more conversations, and access to Reuters’ core financialinformation packages, the Reuters Integrated Data Network(IDN). The new Reuters system still requires a separate terminalto trade rather than allowing access through a trader’s PC and isnot Web-capable.

MARKET MAKERS

A market maker is a person or a firm that quotes a buy and sell pricein a financial instrument or commodity hoping to make a profiton the turn or the spread between the bid price and the ask price.

Forex market makers ensure that the market is always functionaland that the currencies in it will always return the market rate.They do so by updating their prices at very small time intervals andundertaking to trade if requested. Forex market makers must ful-fill their obligations irrespective of whether the economic situationis favorable or unfavorable or whether they lose or profit from thetransaction.

Typical Forex market makers include Gain Capital, CMS Forex,Forex Capital Markets (FXCM), and Global Forex Trading, all ofwhich are regulated by the Commodity Futures Trading Commission(CFTC) of the United States. Another prominent Forex marketmaker is Saxo Bank, which is regulated by the Financial ServicesAuthority (FSA) of Denmark.

Until recently, central banks, commercial banks, and investmentbanks dominated the Forex market. Owing to the entry of Forexmarket makers, other market players such as international moneybrokers, large multinational companies, registered dealers, globalmoney managers, and private speculators have entered the mar-ket in large numbers.

Figure 6-1 illustrates the order-execution mechanism from thevantage point of an independent Forex client.

ORDER FLOW

Order flow is a concept from microstructure finance that refers tosigned volume of trading. Trades can be signed arithmetically aspositive or negative depending on whether the “aggressor” is buy-

Electronic Broking 33

Client Market Maker Interdealer Bank Electronic Broker

Figure 6-1 Order-execution mechanism.

ing or selling. The aggressor is the client who initiates the trade,whereas the dealer who posted the bid/ask quotes is consideredto be on the passive side of the trade. For example, a sale of 1,000units by the trader creates an order flow of -1,000 with the dealerin the underlying currency pair.

In the commodity markets, each futures contract has a stan-dardized unit size and must be executed with one buyer and oneseller simultaneously, thus ensuring a state of equilibrium in themarket. Forex market makers are not bound by the constraint thateach sale in one currency pair must have an equal and oppositecounterpart, a purchase of the same quantity of units. If order flowbecomes too lopsided and deviates significantly from the desiredgoal of order flow equaling zero, market makers can initiate a non-client order with the interbank dealer or the electronic broker ifthey deal direct. It is this transactional process that can influenceprice fluctuations at the lowest level, the electronic broker.

LIVE BOOK

Unfortunately, owing to the lack of centralization of spot currencymarkets, there is no one single order-flow book whereby tradershave ready access to order flow, volume, and open interest infor-mation. Figures 6-2 and 6-3 provide two examples of how orderflow works within the commodity futures framework of theChicago Board of Trade.

The abbreviation ZGZ6 is the electronic trading symbol for agold contract (ZG) with a delivery month of December 2006 (Z6).The contract specification is for 100 troy ounces of fine gold (atleast 0.995 pure). The minimum tick size is 10 cents equivalent to$10 per contract.

Note that buy orders are represented in the upper box as thetwo columns on the left, and sell orders are on the right. Buy pricesare listed in descending order, whereas sell prices are listed inascending order. The difference between the top prices in the twocolumns defines the bid/offer spread, which varies only 1 or 2 ticks

34 Part 2: Exchange-Rate Determinants

in these examples. The contract live book summary is displayed inthe bottom box.

The time lapse between these two live-book examples is only 44seconds yet this is a sufficient duration to increment the tradingvolume by 111 contracts. Access to the analogous live book for spot-currency prices would be a tremendous advantage to spot-currencyspeculators, but sadly, this resource is not presently available tospot-currency traders.

Electronic Broking 35

BUY

4

2

6

57

41

22

13

39

Symbol

Last 3

Last 2

Last 1

Change

Previous Settle

Bid

Ask

Open

High

Low

Today’s Volume

ZGZ6

587.9

587.8

587.8

+1.1

586.7

4 x 587.9

13 x 588.0

582.7

589.4

580.2

29555

587.9

587.8

587.7

587.6

587.5

587.4

587.3

587.2

13

10

39

10

7

8

2

25

588.0

588.1

588.2

588.3

588.4

588.5

588.6

588.7

QTY

Sep 21, 2006 11:01:43 AM

ORDERS

PRICE

SELL

QTY

ORDERS

PRICE

Figure 6-2 Live book and summary, September 21, 2006, 11:01:43 a.m.

To view the CBOT live book in action, traders are directed tothe following Web page: www.cbot.com/cbot/pub/page/0,3181,1069,00.html.


BUY

1

6

30

3

4

51

4

66

Symbol

Last 3

Last 2

Last 1

Change

Previous Settle

Bid

Ask

Open

High

Low

Today’s Volume

ZGZ6

588.4

588.3

588.3

+1.6

586.7

1 x 588.3

14 x 588.6

582.7

589.4

580.2

29666

588.3

588.2

588.1

588.0

587.9

587.8

587.7

587.6

1

14

28

16

29

15

2

24

588.5

588.6

588.7

588.8

588.9

589.0

589.1

589.2

QTY

Sep 21, 2006 11:02:27 AM

ORDERS

PRICE

SELL

QTY

ORDERS

PRICE

Figure 6-3 Live book and summary, September 21, 2006, 11:02:27 a.m.

www.cbot.com/cbot/pub/page/0,3181,1069,00.html

www.cbot.com/cbot/pub/page/0,3181,1069,00.html

Chapter 7Scheduled News

ReleasesNEWS SOURCES

Table 7-1 represents the major agencies that supply economic andfundamental data on a periodic schedule.

37

TABLE 7-1 Major News Sources

Agency Department Web Site

Federal Reserve System U.S. central bank www.federalreserve.gov

Bureau of Labor Statistics Department of Labor www.bls.gov

Bureau of Economic Analysis Department of Commerce www.bea.gov

Bureau of the Census Department of Commerce www.census.gov

Economic Research Service Department of Agriculture www.ers.usda.gov

FedStats Over 100 U.S. federal www.fedstats.govagencies

Statistics Bank of England www.bankofengland.co.uk

National Statistics Official U.K. statistics www.statistics.gov.uk

Statistics Canada National Statistical Agency www.statcan.ca


www.federalreserve.gov

www.bls.gov

www.bea.gov

www.census.gov

www.ers.usda.gov

www.fedstats.gov

www.bankofengland.co.uk

www.statistics.gov.uk

www.statcan.ca


DOMESTIC ECONOMIC CALENDAR (SEE TABLE 7-2)

TABLE 7-2 Domestic Economic Calendar

Economic Release Release Time (ET)

Gross domestic product 8:30 a.m.Consumer price index 8:30 a.m.Employment cost index 8:30 a.m.Producer price index 8:30 a.m.Productivity and costs 8:30 a.m.Employment 8:30 a.m.Personal income 8:30 a.m.Business inventories 8:30 a.m.Durable goods 8:30 a.m.Retail sales 8:30 a.m.Trade balance 8:30 a.m.Housing starts 8:30 a.m.Production and capacity utilization 9:15 a.m.Leading indicators 10:00 a.m.Consumer confidence 10:00 a.m.Wholesale inventories 10:00 a.m.Philadelphia Fed survey 10:00 a.m.Existing home sales 10:00 a.m.Construction spending 10:00 a.m.Report on business 10:00 a.m.Nonmanufacturing report on business 10:00 a.m.New home sales 10:00 a.m.Chicago Fed national activity index 10:00 a.m.Factory orders 10:00 a.m.Federal budget 2:00 p.m.Consumer credit 3:00 p.m.

FOREIGN ECONOMIC CALENDAR (SEE TABLE 7-3)

TABLE 7-3 Foreign Economic Calendar

Market Open Releases Close

Frankfurt 0200 0245 1100London 0300 0430 1200New York 0800 0830 1600Wellington 1500 1530 2300Sidney 1700 1730 0100Tokyo 1800 1850 0200Hong Kong 1900 1930 0300

USAGE

Since economic indicators gauge a country’s economic state,changes in the conditions reported therefore will directly affect theprice and volume of a country’s currency. It is important to keepin mind, however, that the indicators just listed are not the onlythings that affect a currency’s price. There are third-party reports,technical factors, and many other things that also can affect a cur-rency’s valuation significantly.

It is always prudent to check the economic calendars, bothdomestic and foreign, before beginning a new trading session. Twocritical factors concerning news announcements are the magnitudeof impact that the release exerts on exchange rates and the responsetime. Response time can be immediate, usually creating highly vis-ible price surges within the first minute. Other response times maybe delayed or drawn out over the next half hour or so and exhibitless dramatic vertical surging.

Scheduled News Releases 39

PART 3Activity and

Range


Chapter 8Activity

OVERVIEW

Before delving into the complexities of wave theory and shockwaveanalysis, it is expedient to review some of the technical tools avail-able to spot currency traders.

Traders who specialize in currency futures have an advantageover those who trade in spot currencies. This informational edgeconsists of two additional data fields—volume and open interest.The decentralized nature of the foreign exchange markets pre-cludes ready access to these fields except at the very lowest-ordertransaction and accounting levels.

Streaming spot currency data consist of only three fields—date,time, and price. These data can be massaged easily into equispacedinterval data such as 1-minute, 30-minute, hourly, daily, and so on.Historical interval data frequently are available to clients throughthe trader’s online currency platform.

The independent data vendor with the largest and most com-prehensive selection of historical spot currency data is DiskTrading, Ltd. (www.disktrading.com). DTL packages historicalquotes going back to 1972 as date, time, open, high, low, close,upticks, and downticks. An uptick is recorded when a spot priceis greater than its immediate predecessor, and a downtick isrecorded when a spot price is lower than the previous quote.

43


www.disktrading.com

DEFINITION

Given the two fields upticks and downticks, I define activity as follows:

Activityx � upticksx � downticksx

where x is the array index in the time series.Activity is displayed in the lower portions of Figures 8-1 and 8-2

as vertical bars. The light rectangles are upticks, whereas the shadedrectangles are downticks.

Note that activity in the EURUSD currency pair has nearlytripled since January 1, 2000.

EQUIACTIVITY CHART

The equiactivity chart is my brainchild, and I wanted to incorporatethe activity property directly into the OHLC bar chart rather thandisplaying activity at the bottom of the chart as an afterthought. Thisconcept is similar to that of analyst and author Richard W. Arms,

44 Part 3: Activity and Range

1.81111.81011.80911.80811.80711.80611.80511.80411.80311.80211.80111.80011.29911.29811.29711.29611.29511.2941

900850800750700650600550500450400350300250200150100500

GBPUSD-Tue 3/15/20045-Minute OHLC and Activity Chart

0100 0200 0300 0400 0500 0600 0700 0800 0900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000 2100 2200 2300

Figure 8-1 Activity expressed as vertical bars.

Jr., who in the early 1970s introduced the equivolume chart in hisbook Volume Cycles in the Stock Market.

The basic principle is that since the height and range of each ver-tical bar define the high and low prices for that interval, the widthof the vertical bar can be used to represent the activity for the sameinterval. The chart in Figure 8-3 displays the raw data as a conven-tional OHLC vertical bar chart with equal spacing along the x axis.

The chart in Figure 8-4 displays the accordionlike property ofthe equiactivity chart. Column widths increase proportionatelywith an increase in activity during that interval.

COMPARATIVE ACTIVITY

The concept of activity was employed as a means to evaluate theintrinsic characteristics of a specific currency pair. Activity also canbe used to rank multiple currency pairs.

Activity 45

1.38001.36001.34001.32001.30001.28001.26001.24001.22001.20001.18001.16001.14001.12001.10001.08001.06001.04001.02001.0000.9800.9600.9400.9200.9000.8800.8600.8400.8200

200000018000001600000140000012000001000000800000600000400000200000J FMAMJ J A SOND J FMAMJ J A SOND J FMAMJ J A SOND J FMAMJ J A SOND J FMAMJ J A SOND

EURUSD 1/1/2000 to 12/31/2004 Monthly OHLC and Activity Chart

Figure 8-2 Long-term activity.


EURUSD 10/29/2004 Friday 0815 to 09001-Minute Equi-Activity Bar Chart

1.27751.27731.27711.27691.27671.27651.27631.27611.27591.27571.27551.27531.27511.27491.27471.27451.27431.27411.27391.27371.27351.27331.27311.27291.27271.27251.2723

0815 0820 0825 0830 0835 0840 0845 0850 0855 0900

Figure 8-4 Corresponding equiactivity chart.

1.2775

EURUSD 10/29/2004 Friday 0815 to 09001-Minute OHLC Bar Chart

1.27731.27711.27691.27671.27651.27631.27611.27591.27571.27551.27531.27511.27491.27471.27451.27431.27411.27391.27371.27351.27331.27311.27291.27271.27251.2723

0815 0820 0825 0830 0835 0840 0845 0850 0855 0900

Figure 8-3 Conventional OHLC bar chart.

The totals shown in Table 8-1 were compiled from hourly datasupplied by Disk Trading, Ltd. The uptick and downtick data forall currency pairs are based on the date range from January 1, 2000through December 31, 2003.

This table reveals some interesting insights into the currencymarkets and dispels many common misconceptions. Although, asmentioned earlier, trading activity is not an exact match for trad-ing volume, the table does illustrate how actively traded a currencypair is. We know which markets have the most ticks, but we do notknow the size of each trade (a necessary component used to cal-culate trading volume in the commodity futures markets).

First, the top position held by the EURUSD is no surprise.However, the order of the next three pairs is somewhat surprising.One would expect the reverse order: GBP, JPY, and then CHF.

Activity 47

TABLE 8-1 Majors, Minors, and Cross-Rates Sorted by Activity

Rank Pair Upticks Downticks Activity Percent

1 EURUSD 11,054,097 11,030,492 22,084,589 11.072 USDCHF 9,721,359 10,212,108 19,933,467 9.973 USDJPY 9,561,763 9,560,629 19,122,392 9.574 GBPUSD 8,008,837 8,187,172 16,196,009 8.125 EURJPY 7,797,366 7,813,951 15,611,317 7.826 EURCHF 6,607,172 6,652,047 13,259,219 6.667 EURGBP 6,196,830 6,155,652 12,352,482 6.218 GBPJPY 6,137,329 6,040,829 12,178,158 6.119 GBPCHF 5,499,563 5,566,750 11,066,313 5.56

10 USDSEK 5,389,366 5,345,511 10,734,877 5.3611 USDNOK 5,176,038 5,132,495 10,308,533 5.1612 USDDKK 3,834,671 3,768,339 7,603,010 3.8113 USDCAD 3,037,610 3,095,395 6,133,005 3.0614 AUDUSD 2,772,073 2,739,711 5,511,784 2.7615 USDHUF 2,240,576 2,220,757 4,461,333 2.2516 NZDUSD 1,602,812 1,617,497 3,220,309 1.6017 AUDJPY 1,353,009 1,346,705 2,699,714 1.3518 USDCZK 1,186,874 1,176,489 2,363,363 1.2019 CHFJPY 1,104,089 1,137,548 2,241,637 1.1020 USDZAR 652,792 681,038 1,333,830 0.6521 USDSGD 616,208 589,833 1,206,041 0.60

The position of the Scandinavian pairs (SEK, NOK, and DKK)above both the CAD and AUD is also unexpected.

Two former Soviet-satellite currencies (HUF and CZK) haveinched their way up the ladder since the early 1990s and the dis-solution of Soviet Communism.

The low position of the South African rand (ZAR) is difficultto explain based on its wealth of gold and diamonds in inter-national trade.

Lastly, the positions of the cross-rates are slightly higher than antic-ipated. Of course, all these observations are based on the unfoundedbias that the USD is the universal currency around the world.

SINGLE-CURRENCY ACTIVITY

The preceding analysis is based on currency pairs. It is possible toestimate the activity of each individual major currency by summingonly the cross-rate pairs in which it occurs (see Table 8-2).

These are only rough approximations because minor currencypairs were not included in the sampling, nor do these approximationsreflect the volume of each trade. Nonetheless, these examples shouldgive traders a better understanding of Forex activity in general.

DIRECTION

The appellation direction appears throughout the history of techni-cal analysis, and my use of this moniker differs slightly from previ-ous connotations for lack of a better term. Within the context of theanalysis of intrainterval tick data, I will define direction as the differ-ence between the sum of the upticks and the sum of the downticks:


TABLE 8-2 Single Major Currencies Sorted by Activity

Currency Components Activity Percent

USD EURUSD � USDJPY � USDCHF � GBPUSD 77,336,457 27.9EUR EURUSD � EURCHF � EURJPY � EURGBP 63,307,607 22.9GBP USDGBP � EURGBP � GBPJPY � GBPCHF 51,792,962 18.7JPY USDJPY � EURJPY � GBPJPY � CHFJPY 49,153,504 18.7CHF USDCHF � EURCHF � CHFJPY � GBPCHF 35,434,323 12.8

Directionx � upticksx – downticksx

where x is the array index in the time series.There is no direct correlation between my arbitrary definition

of direction and that of J. Welles Wilder, the noted trader/authorof the 1970s who developed the average directional index (ADX).ADX uses a positive directional indicator and a negative directionalindicator to evaluate the strength of a trend. Further informationon Wilder’s method can be found in his highly acclaimed work,New Concepts in Technical Trading Systems (Trend Research, 1978).

Direction can be plotted as a scaled oscillator that ranges from�100 to �100 by dividing the raw direction difference by the activ-ity of the corresponding interval and then multiplying by 100:

Direction oscillatorx � 100(upticksx – downticksx)/(upticksx � downticksx)

where x is the array index in the time series (see Figure 8-5).

Activity 49

0820 0830 0840 0850 0900 0910 0920 0930 0940 0950 1000 1010 1020

1.20901.20871.20841.20811.20781.20751.20721.20691.20661.20631.20601.20571.20541.20511.20481.20451.20421.2039

20

15

10

5

0

-5

-10

-15

-20

EURUSD 1/13/2006 Fri 0820 to 1020 ET 1-Minute OHLC with 1-Minute Direction

Figure 8-5 Direction oscillator chart.

The lower-right vertical scale in this chart is expressed as a percentage of the total possible number of upticks and downticksfor that one interval.

OBSERVATION

Any minute piece of technical information may prove to be valu-able to the spot currency trader regardless of how insignificant orcryptic it may appear. Although activity is not an exact match oftrading volume, it does provide traders with a viable substitute.


Chapter 9Composite

Activity Charts

TIME OF DAY

The Forex is a 24/7 global market (with reduced liquidity onSaturdays and Sundays, of course). The importance of studyingactivity is extremely useful for traders in determining when toschedule their trading sessions. For this purpose, I developed twocomposite charting techniques to show traders when tradingactivity is at its highest and lowest for each currency pair: the time-of-day chart and the day-of-the-week chart. Composite charts are sim-ply the average activity for each time interval sampled over a longtime frame.

In the composite chart in Figure 9-1, multiple time intervalshave been plotted in which each average value has been centered.That is, the 3-minute average for 10:00 a.m. is the mean of the activ-ity for 9:59 a.m., 10:00 a.m., and 10:01 a.m. rather than front-basedaveraging (9:58 a.m., 9:59 a.m., and 10:00 a.m.).

As expected, the highest peak during the 24-hour period is the8:30 a.m. ET spike coinciding with scheduled news announce-ments. The gradual rise in activity between 2:00 and 3:00 a.m.marks the opening of the European markets.

51


DAY OF WEEK

The preceding chart is excellent for examining activity over a sin-gle 24-hour period. For the purpose of scrutinizing activity over anentire week, I created the day-of-week chart, which is simply theconcatenation of six time-of-day charts (see Figure 9-2).

Because of the increased time frame, I also increased the timeinterval (i.e., from 1 minute to 1 hour). Because New York City isconventionally deemed the global center for currency trading, thebottom time scale of all time-of-day and day-of-week charts isexpressed in terms of Eastern U.S. Time (ET) or Greenwich MeanTime (GMT) minus 5 hours.

Several time-of-day and day-of-week charts for the most fre-quently traded currency pairs appear in the appendices. Thesecharts have been updated to mirror the prevailing market char-acteristics and are intended as a computer-side reference guidewhile traders are working in their online currency platforms.


500

475

450

425

400

375

350

325

300

275

250

225

200

175

150

125

100

75

50

25

0

0000 0100 0200 0300 0400 0500 0600 0700 0800 0900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000 2100 2200 2300 2400

EURUSD Time-of-Day Chart (Wednesdays)1-, 3- and 5-Minute Activity (1/1/2004 to 12/31/2004)

Figure 9-1 Time-of-day activity chart.

Composite Activity Charts 53

600057505500525050004750450042504000375035003250300027502500225020001750150012501000750500250

0

Sun Mon Tue Wed Thu Fri

EURUSD Day of Week1-Hour Activity (1/1/2004 to 12/31/2004)

Figure 9-2 Day-of-Week activity chart.

Chapter 10Range

ABSOLUTE RANGE

Traditionally, range has been defined as the difference between thehighest high and the lowest low in a time series over a selectedperiod of time:

Rangex � highest highx – lowest lowx

where x defines an interval in the time series array.The study of range always has provided technical analysts with

vital information on the volatility and trending properties of a timeseries. In fact, without adequate range, speculation in any marketbecomes rather futile.

Graphically, there is nothing very exotic about a range bar chart.It is simply a vertical bar chart of the OHLC quotes at the top ofthe chart, whereas range is also represented as vertical columns atthe bottom of the chart (see Figure 10-1).

Three different range intervals are displayed on this chart: 5-minute, 30-minute, and 60-minute. The vertical scale in thelower-right portion is expressed in terms of pips in the quote(rightmost) currency of the underlying currency pair. Close exam-ination of the vertical bars will reveal the timing of trending cyclesand lateral congestion in the corresponding OHLC chart.

55


RELATIVE RANGE

Relative range is defined as 100 times the quotient of the absoluterange divided by the midrange (the mean of the high and low):

Relative rangex � 100(highx – lowx)/[(highx � lowx)/2]

When examining the properties of a single currency pair, thetrader should use absolute range. When comparing two or morecurrency pairs at the same time, the trader should use relativerange. Absolute range is measured in pips of the quote (second)currency in the currency pair, which may vary radically accordingto the underlying parity rate (e.g., 210 HUF � 1 USD). On theother hand, relative range is a ratio in which the effects of exag-gerated parity rates are automatically filtered out.

Tables 10-1 and 10-2 illustrate how relative range is a better yard-stick for comparing currency pairs than absolute range.


GBPUSD Wednesday 3/31/2004 0000 to 1200 ET5-Min OHLC with 5-Min, 30-Min, 60-Min Range

1.84051.84001.83951.83901.83851.83801.83751.83701.83651.83601.83551.83501.83451.83401.83351.83301.83251.8320

60

50

40

30

20

10

0

0000 0100 0200 0300 0400 0500 0600 0700 0800 0900 1000 1100 1200

Figure 10-1 OHLC with range bar chart.

Range 57

TABLE 10-1 Relative Range, January 1, 2004 to March 31, 2004

Rank Pair High Low Absolute Range Relative Range

1 USDZAR 7.3603 6.2800 1.0803 15.83982 NZDUSD 0.7067 0.6337 0.0730 10.89233 AUDUSD 0.7976 0.7253 0.0723 9.49504 GBPJPY 207.1300 190.3600 16.7700 8.43795 EURJPY 138.7400 127.5400 11.2000 8.41226 AUDJPY 85.1900 78.4800 6.7100 8.19947 USDSEK 7.6599 7.0833 0.5766 7.82198 GBPCHF 2.3865 2.2102 0.1763 7.67079 USDNOK 7.1433 6.6306 0.5127 7.4445

10 USDJPY 112.1400 104.3700 7.7700 7.177511 USDCZK 27.2030 25.3310 1.8720 7.126812 CHFJPY 87.9200 81.9300 5.9900 7.053313 GBPUSD 1.9085 1.7803 0.1282 6.950814 USDCHF 1.2998 1.2200 0.0798 6.333815 EURGBP 0.7060 0.6631 0.0429 6.266916 USDDKK 6.1622 5.7893 0.3729 6.240217 EURUSD 1.2858 1.2081 0.0777 6.231218 USDCAD 1.3477 1.2699 0.0778 5.944419 USDHUF 214.5400 202.3100 12.2300 5.867820 USDSGD 1.7163 1.6704 0.0459 2.710621 EURCHF 1.5833 1.5478 0.0355 2.2676

TABLE 10-2 Relative Range, April 1, 2004 to June 30, 2004

Rank Pair High Low Absolute Range Relative Range

1 USDZAR 7.0599 6.1388 0.9211 13.95742 AUDUSD 0.7671 0.6809 0.0862 11.90613 NZDUSD 0.6671 0.5952 0.0719 11.39194 CHFJPY 89.7300 80.7300 9.0000 10.55975 USDJPY 114.7300 104.0700 10.6600 9.74416 AUDJPY 81.2800 74.5100 6.7700 8.69127 USDCZK 27.6590 25.3800 2.2790 8.59378 EURJPY 137.0400 126.2700 10.7700 8.18059 USDHUF 217.0000 201.6400 15.3600 7.3380

10 GBPJPY 204.7500 190.6400 14.1100 7.137311 USDCAD 1.3979 1.3070 0.0909 6.721112 USDCHF 1.3206 1.2351 0.0855 6.690913 GBPCHF 2.3791 2.2529 0.1262 5.449114 GBPUSD 1.8549 1.7566 0.0983 5.443715 USDNOK 7.0045 6.6380 0.3665 5.372916 EURUSD 1.2366 1.1773 0.0593 4.913217 USDDKK 6.3174 6.0207 0.2967 4.809518 USDSEK 7.7650 7.4036 0.3614 4.765119 EURCHF 1.5663 1.5052 0.0611 3.978520 USDSGD 1.7293 1.6623 0.0670 3.950921 EURGBP 0.6801 0.6547 0.0254 3.8058

OBSERVATIONS

When comparing the order of the currency pairs in two quarters,traders will observe some statistical peculiarities—most notably, thedescent of the Swedish krona from seventh place in the first quar-ter to eighteenth place in the second quarter. Also unexpected isthe rise of the CHFJPY cross-rate from twelfth place to fourth place.

Additionally, it is rather strange that the USDZAR currency pair,which has the second least activity of all 21 currency pairs (seeTable 8-1), has the highest relative range. This warrants a special-ized future study.


59

Chapter 11Composite Range

Charts

TIME OF DAY

In a manner analogous to the one described in Chapter 8, rangesampled over a long time frame can be used to create Time of Dayand Day of Week charts. Simple arithmetic averaging is employedto create representative interval range values.

The Time of Day charts in Figures 11-1 and 11-2 were designedby averaging 2-, 7-, and 15-minute range statistics for the timeframe January 1, 2004 through December 31, 2004 for each week-day except Saturday. The vertical scale on the right is expressed interms of USD pips.

The exaggerated spike at 8:30 a.m. ET indicates the time ofweek when the most influential U.S. news announcements arereleased. This means that spot currency traders on the West Coastmust have already consumed the appropriate quantity of coffeeand powered up the trading platform by 5:15 a.m.



20.0

19.0

18.0

17.0

16.0

15.0

14.0

13.0

12.0

11.0

10.0

9.0

8.0

7.0

6.0

5.0

4.0

3.0

2.0

1.0

0.0

0010 0100 0200 0300 0400 0500 0600 0700 0800 0900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000 2100 2200 2300 2400

EURUSD Time of Day Chart (Mondays)2-, 7- and 15-Minute Range (1/1/2004 to 12/31/2004)

Figure 11-1 Time of Day range chart (Mondays).

EURUSD Time of Day Chart (Fridays)2-, 7- and 15-Minute Range (1/1/2004 to 12/31/2004)

0010 0100 0200 0300 0400 0500 0600 0700 0800 0900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000 2100 2200 2300 2400

56.054.052.050.048.046.044.042.040.038.036.034.030.0

28.030.0

26.024.022.020.018.016.014.012.010.08.06.04.02.00.0

Figure 11-2 Time of Day range chart (Fridays)

DAY OF WEEK

In a similar fashion, the Time of Day range charts can be concate-nated to create a Day of Week range chart. The distinguishing timeinterval must be increased accordingly (see Figure 11-3).

AVERAGE INTERVAL RANGE

The study in Figure 11-4 uses 1-minute high/low data from January1, 2000 through December 31, 2003 in the EURUSD currency pair.I calculated the average range (high minus low) for each integertime interval from 1 to 60 minutes.

The average range interval chart provides traders with someinteresting theoretical information. Note on the far left that theheight of the first vertical bar is only 5.14 pips. This means that for every 1-minute interval in the horizontal scaling of the dealer’s

Composite Range Charts 61

EURUSD Day of Week1-Hour Range (1/1/2004 to 12/31/2004)

70686664626058565452504846444240383634323028262422201816141210

86420


Figure 11-3 Day of Week range chart.

trading platform, there is a high statistical expectancy that theprice movement will span 5.14 pips.

Within every half-hour representation in the trading platform,the price movement should cover a range of 18.5 pips. Every hourin the trading window, prices should span 21.2 pips (the rightmostvertical bar). Of course, traders must compare this informationwith the corresponding Time of Day range chart before enteringthe market based solely of range indicators.


EURUSD - 1/1/2000 to 12/31/2003Average Interval Range (1-Minute Increments)

25242322212019181716151413

1112

109876543210

0 5 10 15 20 25 30 35 40 45 50 55 60

Figure 11-4 Average range interval chart.

63

Chapter 12Activity versus

Range

FORECASTING MODEL

This chapter will examine the dominant influence that activity has on range. This is to say, I will assign activity as the independentvariable and range as the resulting dependent variable in a simplelinear model:

Range � A(activity) � B

where A equals the slope and B equals the intercept of the regres-sion line.

SCATTER DIAGRAMS

Figures 12-1 through 12-4 are scatter diagrams for four differenttime periods, all expressed in Eastern Time. The vertical scale onthe left is range expressed in pips, and the horizontal scale at thebottom is activity expressed in ticks. All diagrams cover the samedate range of January 1, 2006 through March 31, 2006.



EURUSD 1/1/2006 to 3/31/2006 0001 to 2359Activity (x-axis) versus Range (y-axis)

Slope = .0191 Intercept = 2.89 Cor = 65.64

45

40

35

30

25

20

15

10

5

0 50 100 150 200 250 300 350 400 450 500 550 600 650 700

Figure 12-1 Time frame 0001 to 2359 ET.

45

40

35

30

25

20

15

10

5

0 50 100 150 200 250 300 350 400 450 500 550 600




Activity versus Range 65

45

40

35

30

25

20

15

10

5

0 50 100 150 200 250 300 350 400 450 500 550 600




45

40

35

30

25

20

15

10

5

0 50 100 150 200 250 300 350 400 450 500 550 600




REGRESSION RESULTS

The preceding four charts generated the data shown in Table 12-1.As expected logically, activity shows itself to be an adequate inde-

pendent variable to estimate range over the same time frame. Infact, activity may very well be the only valid estimator for rangeexcept for some possible serial correlations.

An interesting statistical point is that as the time span decreases,both the slope and the coefficient of correlation increase, albeitvery gradually. This is so because the diminishing time frames elim-inate more and more periods of low volatility, where range tendsto approach zero.

USAGE

When traders have access to the number of ticks in the streamingdata for the time frame 8:30 a.m. to 10:00 a.m., 1-minute range canbe forecast with the linear model:

Range � 0.0201(number of ticks) � 2.96

Even though a range estimate does not directly assist in fore-casting price direction, it still can be instrumental in determiningthe placement of stop-loss and take-profit limit orders. Of course,all this will occur in real time, and scalpers must be aware of therisks involved.


TABLE 12-1 Activity versus Range Correlation

Time Span Slope Intercept Correlation

0001 to 2359 0.0191 2.89 65.640830 to 1200 0.0193 3.00 66.030830 to 1000 0.0201 2.96 68.390830 to 0845 0.0237 2.50 71.87

Chapter 13Inertia

OVERVIEW

Before introducing the concept of economic inertia, I first mustreview a common technical tool called momentum. Traditionally,momentum has been defined as the difference between the currentclosing price and a previous closing price:

Momentumx � closex – closex – lag

where x is the day number and lag is the number of elapsed days.I mention this because Forex inertia has a very direct relationshipwith momentum that is explained below.

RAW CURRENCY DATA

Historical data for spot currency pairs can be obtained from a vari-ety of data suppliers. One such purveyor is Disk Trading, Ltd.(www.disktrading.com), which currently offers data for 23 majors,minors, and cross-rates. The data can be obtained as comma-sep-arated files (CSV) in either of two formats—streaming tick data orinterval data.

Packaged tick data use the format Date, Time, Price. Adjacentprices in the streaming tick data necessarily will have the same time

67


www.disktrading.com

field unless trading volume has dropped sharply or a full minutehas elapsed. Up to 700 ticks have been observed in a single minute.This occurred at 8:31 a.m. ET in the EURUSD currency pair on aFriday.

Interval data are simply streaming tick data that have been mas-saged into standard equispaced time intervals such as 1-minute, 10-minute, 30-minute, hourly, and daily. During the “intervaliza-tion” process of streaming data, the number of upticks anddownticks per interval are each calculated. Thus historical inter-val data come packaged using the following format: Date, Time,Open, High, Low, Close, Upticks, and Downticks for each time inter-val. The uptick tally is incremented if the streaming tick price isgreater than its immediate predecessor, and a new downtick isrecorded if the streaming tick price is less than its predecessor.

In both tick and interval data, dates are expressed as mm/dd/yyyy,and time is represented as hhmm using the 24-hour convention withno intervening punctuation

FOREX INERTIA

When traders have access to historical streaming tick data, anotherstatistic is available with the aid of a little arithmetic. Activity sim-ply “counts” the number of upticks and downticks during a singletime unit. Inertia, on the other hand, measures the magnitude ofupticks and downticks. Thus I define (albeit somewhat arbitrarily)inertia as follows:

Inertiax � sum(positive differentialsx) � sum(negative differentialsx)

Note that the absolute values of the sum of the negative differ-entials are used. Assume that we have the sample streaming tickdata shown in Table 13-1.

These raw tick data can be coerced to reveal additional intrain-terval information as shown in Table 13-2.


Inertia 69

TABLE 13-1 Raw Currency Tick Data

Date Time Price

10/11/2005 1009 12344

10/11/2005 1010 12346

10/11/2005 1010 12348

10/11/2005 1010 12343

10/11/2005 1011 12349

10/11/2005 1011 12352

10/11/2005 1011 12350

10/11/2005 1011 12355

10/11/2005 1011 12356

10/11/2005 1012 12360

10/11/2005 1012 12359

10/11/2005 1012 12365

10/11/2005 1012 12358

10/11/2005 1012 12354

TABLE 13-2 Intrainterval Data

Up- Down- Up DownTime Price ticks ticks Activity Direction Inertia Inertia Inertia Momentum

1009 12344 0 0 0 0 0 0 0 0

1010 12346

1010 12348

1010 12343 2 1 3 1 4 5 9 �1

1011 12349

1011 12352

1011 12350

1011 12355

1011 12356 4 1 5 3 15 2 17 �13

1012 12360

1012 12359

1012 12365

1012 12358

1012 12354 2 3 5 �1 10 12 22 �2


The sum of the up inertia and the down inertia provides thetrader with another volatility indicator that, for all intents and pur-poses, is even more informative than the activity indicator becausethe magnitude of each price fluctuation is included.

It also should be noted that the difference between up inertiaand down inertia is equivalent to the momentum of that timeinterval.

INERTIA CHARTS

The difference between an activity chart and an inertia chart ofthe same data is displayed in Figures 13-1 and 13-2.

A second example of the difference between activity and iner-tia is shown in Figures 13-3 and 13-4.

1.20901.20871.20841.20811.20781.20751.20721.20691.20661.20631.20601.20571.20541.20511.20481.20451.20421.2039

300

350

250

200

150

100

50

0

0820 0830 0840 0850 0900 0910 0920 0930 0940 0950 1000 1010 1020

EURUSD 1/13/2006 Fri 0820 to 1020 ET 1-Minute OHLC with 1-Minute Activity

Figure 13-1 OHLC bar chart with activity oscillator.

Inertia 71

1.2088

1.2083

1.2078

1.2073

1.2068

1.2063

1.2058

1.2053

1.2048

1.2043

1.2038

600525450375300225150750

0820 0830 0840 0850 0900 0910 0920 0930 0940 0950 1000 1010 1020

EURUSD 1/13/2006 Fri 0820 to 1020 ET1-Minute OHLC with 1-Minute Inertia

Figure 13-2 OHLC bar chart with inertia oscillator.

1.20251.20201.20151.20101.20051.20001.19951.19901.19851.19801.19751.19701.19651.19601.19551.1950

350

300

250

200

150

100

50

0

0820 0830 0840 0850 0900 0910 0920 0930 0940 0950 1000 1010 1020

EURUSD 3/24/2006 Fri 0820 to 1020 ET1-Minute OHLC with 1-Minute Activity

Figure 13-3 Second example (activity).

OBSERVATIONS

The visible property that we can observe in the preceding exam-ples is that the inertia oscillator produces higher highs and lowerlows than the corresponding activity oscillator.

To illustrate the greater “accuracy” of inertia over activity, weshould examine the following anomaly: Assume that during thespan of 1 minute there are six upticks each with a magnitude of 1pip. There is also one downtick during the same interval, and ithas a magnitude of –12 pips. Therefore, activity is recorded as 7ticks total, whereas combined inertia has a value of �18.

MOMENTUM VERSUS INERTIA

Combined inertia (the sum of up inertia and down inertia) can bedefined more accurately as the sum of the absolute values of theintrainterval tick differentials. In this instance, a differential is the


1.20251.20201.20151.20101.20051.20001.19951.19901.19851.19801.19751.19701.19651.19601.19551.1950

75067560052545037530022515075

0

0820 0830 0840 0850 0900 0910 0920 0930 0940 0950 1000 1010 1020

EURUSD 3/24/2006 Fri 0820 to 1020 ET 1-Minute OHLC with 1-Minute Inertia

Figure 13-4 Second example (inertia).

difference between the current tick price and the immediately preceding tick price.

As mentioned earlier, the difference between the up inertia sumand the down inertia sum renders the momentum for that inter-val in question. This gives rise to another very plausible oscillator—momentum divided by inertia. This statistic may be indicative ofcertain properties that occur at the order-flow level and is currentlyin the testing phase.

Lastly, I am aware that within the realm of the physical sciencesthe formal definition of inertia is the property of matter by whichit remains at rest or in uniform motion in the same straight lineunless acted on by some external force. I arbitrarily selected thisinertia moniker to distinguish it from momentum as it refers totime-series analysis and yet show its close affiliation to time-seriesmomentum.

Inertia 73

PART 4Forex Wave

Theory


Chapter 14Review of Wave

Principles

OVERVIEW

In the preceding chapters I have reviewed numerous fundamentaland technical influences that cause spot currency prices to fluctu-ate. I also have devised several innovative technical tools that canbe employed expeditiously in the endeavor of deciphering theinner workings of the exchange-rate phenomenon.

In the book Forex Wave Theory (New York: McGraw-Hill, 2006),I introduced several innovative ideas for forecasting time-seriesprices that differ dramatically from the prevailing classical theory,Elliott’s wave principle. These new concepts are essentially basedon an alternative method for identifying, categorizing, and label-ing the various wave patterns generated from standard and modi-fied swing reversal algorithms. Statistical and mathematicalmethods then are applied to uncover the identity and calculate themagnitude of subsequent wave patterns.

One central theme of Ralph N. Elliott’s cycle is the wave compo-sition of an impulse cycle and the wave composition of the subse-quent corrective cycle. Both are examined below. Elements of Forexwave theory that are applicable to the study of economic shockwavesare reviewed immediately after the section on Elliott’s ideas.

77


REVERSAL CHARTS

The primary purpose of reversal charts is to filter out minor fluctuations in the raw data commonly referred to as white noise.In turn, this process accentuates the local extremes, maxima andminima. Probably the most commonly employed reversal chart isthe point and figure chart, which dates back to the late nineteenthcentury. Price movements are represented as columns of X’s(uptrends) and O’s (downtrends) as shown in Figure 14-1.

The point and figure (P&F) chart is somewhat unique withinthe realm of occidental charting systems in that it distorts the timeelement along the x axis. That is, a new column is not appendedto the right of the current column until a minimum reversalamount in the opposite price direction has been satisfied.

On the other hand, a swing reversal chart represents the samereversal data as a series of interconnected diagonal straight lines

78 Part 4: Forex Wave Theory

1.2246

X

0

0 0000

0 00 0

00 0

000000000 0 0 0 0

000 0

0 0 000 0

0

00 0

0 00000000000000 0

000000 0

0000 0

0 000 0

00000

0

00

0 000

000

000000000 0

000

0000

000000000

00

000

000

0

000

00000

00

000

0

00

0000

XX X

X XXXX X

X

XX X

XXX X X

X XXXXX

XXXXXX X

XXX X

XXX X

XXXX X

XXXX X

XX

X XXX X

XXX

XX

XXXXXXXXX X

XX X

XXXX X

XX

XX

X

XXX

XX

XXXXX

XXXX

XX

XXX

XXXX

X

X XXX

XXXXX

EURUSD Wed 9/15/04 07:30 to 10:00 AM ITP&F Chart Box Size - 0.0002 Rev Amt - 3 Boxes

1.22441.22421.22401.22381.22361.22341.22321.22301.22281.22261.22241.22221.22201.22181.22161.22141.22121.22101.22081.22061.22041.22021.22001.21981.21961.21941.21921.21901.21881.21861.21841.21821.21801.21781.2176

Figure 14-1 Point and figure chart.

that can be fitted to overlay the original OHLC data precisely, asin Figure 14-2.

Both camps (the X’s and O’s enthusiasts and the straight-lineaficionados) have their die-hard proponents. For the purpose of Forex wave theory and shockwave analysis, I side heartily withthe straight-liners. Retaining a proper perspective on the x axiswill prove beneficial later and will provide additional informationpertinent to calculating precise shockwave durations along the x axis.

ELLIOTT WAVES

The most widely accepted of all wave forecasting methods isundoubtedly R. N. Elliott’s wave principle, which he developedduring the late 1930s. A Google Internet search on “Elliott waves”produces over 2 million matches.

Review of Wave Principles 79

1.2070

1.2065

1.2055

1.2050

1.2060

1.2045

1.2040

1.2030

1.2025

1.2035

1.2015

1.2010

1.2005

1.2000

200

150

100

50

0

1.2020

EURUSD Wednesday 07/28/2004 08:00-10:00 AM ET1-Min OHLC, G-Box Reversal Swing Chart with 1-Min Activity

Figure 14-2 Swing reversal chart.

RECURRING WAVE PATTERNS

Elliott believed that one business cycle consisted of a five-waveimpulse component (Figure 14-3) and a three-wave corrective com-ponent (Figure 14-4).


12

34

5

Figure 14-3 Elliott impulse pattern.

A

B

C

Figure 14-4 Elliott corrective pattern.

This pattern always satisfies the logical condition

Wave 1 � wave 2 � wave 3 � wave 4 � wave 5

A corrective pattern consists of three waves that represents a coun-tertrend in the opposite direction to its preceding impulse pattern.The corrective pattern must obey the following constraint:

Wave A � wave B � wave C

Elliott identified several other wave patterns (such as zigzags, tri-angles, flats, and so on), but the impulse and corrective patternsare the most important when analyzing trends.

WAVES WITHIN WAVES

Elliott proposed that economic waves exist at many levels, meaningthat there could be waves within waves. Figure 14-5 shows how pri-mary waves could be broken down into smaller component waves.


Figure 14-5 Waves within waves.

1

2

4

A

C

3

5

B

This figure displays four of the eight waves as their componentwaves:

Wave 3 consists of a five-wave bull impulse cycle.

Wave 4 consists of a three-wave corrective cycle.

Wave A consists of a five-wave bear impulse cycle.

Wave B consists of a three-wave corrective cycle.

Elliott assigned the nomenclature shown in Table 14-1 to cyclesin order of descending size.

The concept of waves within other waves dates back to CharlesDow and probably earlier. However, Dow believed that there wereonly three levels of wave composition. The scientific name for thisphenomenon is fractal geometry. The points of interest are the inter-sections between cycles of different fractal levels.

RULES AND GUIDELINES

There are over 273 major rules and guidelines in the Elliott waveprinciple. Elliott rules must be obeyed in detail for a pattern toqualify as an Elliott wave or cycle. The guidelines do not have tobe obeyed precisely. The more guidelines obeyed by an Elliott pat-tern, the higher is its probability of being correct.


TABLE 14-1 Elliott Cycle Names

Rank Cycle Name

1 Grand supercycle2 Supercycle3 Cycle4 Primary5 Intermediate6 Minor7 Minute8 Minuette9 Subminuette

A very brief summary of the rules governing a bull impulse cycleis as follows:

• No part of wave 2 can be more than retrace wave 1.

• Wave 2 must retrace wave 1 by a minimum of 20 percent.

• The maximum time for wave 2 is nine times wave 1.

• Wave 3 must be longer than wave 2 in gross distance by price.

• Wave 3 and wave 1 cannot both have wave 5 failures.

• Wave 3 cannot be less than a third of wave 1 by price.

• Wave 3 cannot be more than seven times wave 1 by price.

• The absolute maximum time limit for wave 3 is seven times wave 1.

A comprehensive list of the major rules and guidelines for theElliott wave principle can be found at www.geocities.com/WallStreet/Exchange/9807/Charts/SP500-Articles/EWRules.htm#imprule.


www.geocities.com/WallStreet/Exchange/9807/Charts/SP500-Articles/EWRules.htm#imprule

www.geocities.com/WallStreet/Exchange/9807/Charts/SP500-Articles/EWRules.htm#imprule

Chapter 15Cycle

Preliminaries

DEFINITIONS

In this study, I prefer to use the original terms of author/theo-retician R. N. Elliott to avoid any unnecessary confusion with termsused by other swing analysts.

A wave is a single straight line in the swing chart. Waves arealways diagonal lines with positive or negative slope, never perfectlyhorizontal or vertical.

A peak is the point of intersection between an upward wave onthe left and a downward wave on the right. This represents a localmaximum in the raw data (see Figure 15-1).

A valley (or trough) is the point of intersection between a down-ward wave on the left and an upward wave on the right. This rep-resents a local minimum in the raw data.

A cycle is a series of adjacent interconnected waves depicting spe-cific price formations.

To convert a sequence of raw tick data or OHLC interval data to its corresponding swing data, a swing reversal algorithm isemployed in which two user-supplied variables must be initialized— the minimum fluctuation unit and the minimum reversal amount.

85


MINIMUM FLUCTUATION UNIT

Traditionally, the minimum price unit is the smallest fractionalprice increment that the quote currency (or underlying security)can change. This is equivalent to the box size used in P&F chart-ing. In the currency markets, this increment is a single pip. Forexample, if the EURUSD currency pair is currently trading at1.2451, a single pip is 0.0001 USD.

There are three cases where a minimum fluctuation unit greaterthan 1 pip might be used. The first is when the parity rate betweentwo currencies is very wide and causes a very large bid/ask spread.For example, if the bid/ask spread (transaction cost) for the EURCZK currency pair is 350 koruny, then a 1-pip box size willhave very negligible filtering power.

A second reason for using a box size greater than 1 pip occurswhen performing historical analysis and a longer time frame isbeing analyzed. In this case, the analyst probably will be scrutiniz-ing major reversals and may have little interest in minor reversals.This pertains more to long-term position traders rather than to ses-sion or day traders.

Lastly, a larger box size may be used to align peaks and valleyswith the grid lines of a chart. This is purely a display preference,though.


Peaks Peaks

Valleys

Figure 15-1 Peaks and valleys.

MINIMUM REVERSAL AMOUNT

The reversal amount is the number of minimum fluctuation unitsnecessary to plot a reversal in price direction. For instance, if thecurrent trend is upward and the reversal amount is set at threeunits, then a decline of three fluctuation units must be reachedbefore the downward movement is plotted. If, instead, a new pricecontinues in the same direction as the existing trend, then singleboxes are added automatically to the last extreme (either a peakor a valley).

It is this mutual interaction between the minimum fluctuationunit and the reversal amount that triggers the reversal mechanismin the swing algorithm necessary to plot peaks and valleys whileignoring lateral price movements.

There is one final case for increasing the minimum fluctuationunit not mentioned earlier. If an analyst, for whatever reasons, hasbecome very partial to one specific reversal amount, it is possibleto increase the minimum fluctuation unit instead of the reversalamount when market conditions change.

For example, a three-unit reversal amount is favored by mosttraders. If traders wish to filter out some of the minor swings, theycan increase either the reversal amount or the minimum fluctua-tion unit. However, keep in mind that though a 2-pip unit size witha three-unit reversal amount algorithm will generate results verysimilar to a 1-pip unit size with a six-unit reversal amount algo-rithm, they will not be identical. This requires some reflection. Thereason is that when you plot a continuation of an existing trend,you can plot smaller distances.

SWING REVERSAL ALGORITHM

The primary purpose of the swing reversal algorithm is to filter outwhite noise. Given the preceding information and user-suppliedvariables, I now will define the swing reversal algorithm as follows(this algorithm assumes that we are using daily OHLC quotes asthe input data rather than simply the closing prices):

Cycle Preliminaries 87

Step 1: Initialize BoxSize and ReversalAmount variables.

Step 2: Create a new variable called Direction.

Step 3: Create two array variables called Price and Time to holdthe swing data.

Step 4: Set Price(1) 5 Close(1) and Time(1) 5 1.

Step 5: If High(2) – Price(1) . BoxSize * ReversalAmount, then

Set Price(2) 5 High(2).

Set Time(2) 5 2.

Set Direction 5 UP.

ElseIf Price(1) – Low(2) . BoxSize * ReversalAmount, then

Set Price(2) 5 Low(2).

Set Time(2) 5 2.

Set Direction 5 DOWN

Else

Increment day number and repeat step 5.

End If

Step 6: Increment DayNo.

If DayNo 5 Number of OHLC quotes, then

Go to step 9.

If Direction 5 DOWN, then

Go to step 8

End If

Step 7: If High(DayNo) – Price(Idx) . BoxSize, then

Set Price(Idx) 5 High(DayNo).

Set Time(Idx) 5 DayNo.

ElseIf Price(Idx) – Low(DayNo) . BoxSize * ReversalAmount, then

Increment Swing Idx.

Set Price(Idx) 5 Low(DayNo).


Set Direction 5 DOWN.


End If

Go to step 6

Step 8: If High(DayNo) – Price(Idx) . BoxSize * ReversalAmount,then

Increment Swing Idx.

Set Price(Idx) 5 High(DayNo).


Set Direction 5 UP.

ElseIf Price(Idx) – Low(DayNo) . BoxSize, then

Set Price(Idx) 5 Low(DayNo).


End If

Go to step 6

Step 9: Set Number of Swings 5 Swing Idx.

Exit

At this point, the two swing arrays Price() and Time() have beenpopulated with corresponding pairs of swing data.

Cycle Preliminaries 89

Chapter 16Basic Three-Wave

Cycles

OVERVIEW

The criterion for identifying a basic three-wave cycle formation isthe relationship between the heights of the individual waves usingthe three comparison operators

� greater than

� equal to

� less than

These operators define nine unique bull cycles and nine uniquebear cycles. If the first wave in any cycle is upward, then the entirecycle is referred to as a bull cycle. If the first wave in any cycle isdownward, then the entire cycle is called a bear cycle. Each three-wave cycle is identified by a one-letter label (ID). Bull cycles usethe uppercase letters A through I, whereas bear cycles are desig-nated by the lowercase letters a through i.

91


IMPULSE CYCLE (ID � A)

The single-most important three-wave cycle is called an impulsecycle, a term borrowed from the Elliott principle. This patterndefines a clear and consistent trend in either price direction, thusindicating an opportunity for profit. The cycle begins with a pricesurge followed by a retracement wave whose height is less than theheight of the initial surge wave. The height of the final wave mustexceed the height of the retracement wave (see Figure 16-1).

RECTANGLE (ID � B)

The second three-wave cycle occurs less frequently than theimpulse cycle because the heights of all three waves must be thesame. The rectangle formation represents a horizontal price move-ment, also called lateral congestion (see Figure 16-2).


Wave 1 > <Wave 2 Wave 3

Figure 16-1 Impulse cycle.

Wave 1 = =Wave 2 Wave 3

Figure 16-2 Rectangle.

CONTRACTING DESCENDING TRIANGLE (ID � C)

This cycle is identified as two waves of the same height followedby one wave whose height is less than either of its two predecessorwaves (see Figure 16-3).

CONTRACTING ASCENDING TRIANGLE (ID � D)

This cycle is identified as a price surge wave followed by two waveswhose heights are equal to and less than the height of the initialwave (see Figure 16-4).

CONTRACTING SYMMETRICAL TRIANGLE (ID � E)

This cycle is identified as an initial price surge wave followed bytwo waves whose heights are less than the immediate predecessorwave. This cycle is very important because a reversal in trend has

Basic Three-Wave Cycles 93

Wave 1 = >Wave 2 Wave 3

Figure 16-3 Contracting descending triangle.

Wave 1 > =Wave 2 Wave 3

Figure 16-4 Contracting ascending triangle.

not occurred at the end point of wave 1 (also the starting point ofwave 2) triggering a downward trend (see Figure 16-5).

EXPANDING ASCENDING TRIANGLE (ID � F)

This cycle is identified as two waves whose heights are equal fol-lowed by a final wave whose height is greater than either of its pre-decessor waves (see Figure 16-6).

EXPANDING DESCENDING TRIANGLE (ID � G)

This cycle is identified as an initial wave whose successor waves areboth greater than the initial wave but equal to each other (seeFigure 16-7).


Wave 1 > >Wave 2 Wave 3

Figure 16-5 Contracting symmetrical triangle.

Wave 1 = <Wave 2 Wave 3

Figure 16-6 Expanding ascending triangle.

EXPANDING SYMMETRICAL TRIANGLE (ID � H)

This cycle is identified as an initial price surge wave followed bytwo waves whose heights are greater than the immediate prede-cessor wave. This cycle is important because it may signal that abear trend is ending and a bull trend is beginning at the end pointof the second wave (see Figure 16-8).

CONNECTOR (ID � I)

This cycle is so named because it links two impulse cycles togetherto create an even longer trend cycle. The height of the middlewave is always greater than the heights of either two adjacent waves(see Figure 16-9).


Wave 1 < =Wave 2 Wave 3

Figure 16-7 Expanding descending triangle.

Wave 1 < <Wave 2 Wave 3

Figure 16-8 Expanding symmetrical triangle.

BEAR CYCLE CONVERSION

Each three-wave bull cycle can be converted to an equivalent bearcycle simply by rotating the cycle 180 degrees along its x axis (orin nontechnical parlance, flipping the cycle vertically). The heightcomparison constraints remain the same, but the first wave nowpoints downward. Four of the triangle designations becomereversed (see Table 16-1).

FREQUENCIES

In the Table 16-2 the frequencies of the three-wave cycles areexpressed as percentages for several minimum reversal amounts.Bull cycles and their complementary bear cycles have been addedtogether to enhance statistical stability.

Several interesting cycle properties are exhibited in this table.As the minimum reversal amount for the swing algorithmincreases, two seemingly paradoxical phenomena occur:

1. The frequencies of the impulse cycles, both symmetrical trian-gles, and the connector cycles also increase.

2. The frequencies for the rectangles, both ascending triangles,and both descending triangles decrease.

The reason for this is quite simple: The cycles that decrease infrequency all have equality relationships between the heights of


Wave 1 < >Wave 2 Wave 3

Figure 16-9 Connector.

two or more waves in those cycles. Naturally, as the heights of twoadjacent waves increase, there is less likelihood that they will bethe same magnitude because the y-axis resolution is growing moreand more gradated. Mathematically, this is analogous to the out-come of throwing different numbers of dice. Tossing two dice atone time renders 36 possibilities. Throwing four dice at the sametime generates 1,296 possibilities (thus greater gradation).

One more curiosity can be observed in this table. In all the dif-ferent minimum reversal amounts, the connector cycle alwaysranked at the top of the list in frequency count, even surpassingthe impulse cycle, which I originally expected to head the list. Thisfact will be useful in later studies.


TABLE 16-1 Bear Cycle Conversions

No. Bull Name ID Bear Name ID

1 Impulse A Impulse a2 Rectangle B Rectangle b3 Contracting Descending Triangle C Contracting Ascending Triangle c4 Contracting Ascending Triangle D Contracting Descending Triangle d5 Contracting Symmetrical Triangle E Contracting Symmetrical Triangle e6 Expanding Ascending Triangle F Expanding Descending Triangle f7 Expanding Descending Triangle G Expanding Ascending Triangle g8 Expanding Symmetrical Triangle H Expanding Symmetrical Triangle h9 Connector I Connector i

TABLE 16-2 Three-Wave Cycle Frequencies

ID Cycle Name 5-Pip 10-Pip 15-Pip 20-Pip 25-Pip 30-Pip 35-Pip

A Impulse 13.3 22.9 29.5 30.9 31.8 32.5 33.7B Rectangle 12.8 3.7 0.6 0.2 0.0 0.1 0.0C Con Desc Triangle 7.6 4.2 1.7 0.8 0.5 0.4 0.3D Con Asc Triangle 13.1 8.6 3.5 2.2 1.0 0.7 0.7E Con Sym Triangle 6.9 11.3 14.5 15.4 15.7 16.1 13.5F Exp Asc Triangle 13.0 8.3 3.6 2.4 1.2 0.7 0.8G Exp Desc Triangle 7.5 3.9 1.8 1.1 0.6 0.4 0.4H Exp Sym Triangle 6.9 9.7 13.6 14.7 16.9 16.2 16.6I Connector 18.9 27.3 31.6 32.2 32.3 32.9 34.0

Chapter 17Forecasting the

Third Wave

OVERVIEW

Financial market forecasts come in many different packages. Whendealing with econometric models (such as Box-Jenkins plots,Fourier transforms, trigonometric regressions, and so on), the esti-mate invariably will be a discrete numeric value or a series of suchvalues. When employing crossover trading systems, the forecast isalmost always given as a market action such as initiate a new longtrade, enter market short, or liquidate any open position. Withinthe confines of Forex wave theory, the forecast is represented as alikelihood relationship.

FORECASTING WAVE RELATIONSHIPS

Table 16-1 not only exhibits the frequency data for three-wavecycles but also provides sufficient information to forecast the logical length of the third wave when only the first two waves areknown. By logical length, I mean the result of the comparative ratiolength using the logical operators �, �, and �.

99


PRACTICAL EXAMPLE

For example, assume that the final two waves in the swing dataform the pattern shown in Figure 17-1.

We automatically know that there are three possible continua-tions once the next wave is known: an impulse cycle, a contract-ing ascending triangle, and a contracting symmetrical triangle (seeFigure 17-2).

From Table 16-1, we obtain and list in Table 17-1 the percent-ages for a 35-pip reversal amount for the possible continuations.

The next step is to sum the percentages in Table 17-1 and dividethat sum into each value (see Table 17-2):

33.7 � 0.7 � 13.5 � 48.9

Given the two-wave pattern in Figure 17-1, this means thatwhen we use a 35-pip swing reversal amount, there is a 69 percentlikelihood that the next wave in the swing data will surpass the topof wave 1 and become a “full” three-wave impulse cycle.


Figure 17-1 Two-wave pattern (wave 1 � wave 2).

Figure 17-2 Possible third-wave continuations.

CAVEAT

Although a 69 percent probability is a rather respectable triggervalue when considering a possible market entry order, the disad-vantage is that the standard deviation is very high when basing theforecast solely on the two preceding wavelengths. However, thereare methods to lower the standard deviation and improve the levelof confidence in later chapters. (This involves basing the forecaston more than just two preceding waves.)

Forecasting the Third Wave 101

TABLE 17-1 Raw Forecast Percentages

Cycle Name Percent

Impulse 33.7Contracting ascending triangle 0.7Contracting symmetrical triangle 13.5

TABLE 17-2 Adjusted Forecast Percentages

Cycle Name Percent

Impulse 69.0Contracting ascending triangle 1.4Contracting symmetrical triangle 27.6

Chapter 18Multiple-Wave

Cycles

OVERVIEW

In earlier chapters I indicated that the basic cycle consists of threeconsecutive waves and is identifiable by a single unique letterbetween A and I or a and i using the three logical operators �, �,and �. In this chapter I will introduce a method for naming cycleswith four or more waves.

OVERLAPPING WAVES

When analyzing cycles consisting of more than three waves, I have devised a naming convention that uses an overlapping wavetechnique.

For example, the four-wave cycle in Figure 18-1 can be decom-posed into the two overlapping three-wave cycles in Figure 18-2.

Wave 2–3 on the left aligns perfectly with wave 2–3 on the right.The same is true for the waves 3–4 on the left and the right. Thethree-wave cycle on the left is a bull impulse cycle (designated A),whereas the three-wave cycle on the right is a bear connector cycle(designated i). Thus, in my naming scheme, the four-wave cyclein Figure 18-1 is labeled Ai.

103


This naming convention can be extended indefinitely by addinga new letter to the current cycle name each time a new wave isappended. In the next example, a six-wave cycle is “decomposed”and labeled (see Figure 18-3).

The individual three-wave components are listed in Table 18-1.Thus the six-wave cycle in Figure 18-3 is identified as an AiFa

cycle pattern.Note that the first and second waves of the rightmost cycle are

overlaid exactly on the second and third waves in the leftmostcycle. Both the heights (y axis) and the widths (x axis) of the twoshared waves must align perfectly to form a new elongated cycle.


1

2

3

4

5

Figure 18-1 Basic four-wave cycle.

1

22

33

4 4

5

Figure 18-2 Four-wave cycle components.

NAMING CONVENTION

It is obvious that the four-wave cycle in Figure 18-1 is actually theconcatenation (or partial merging) of a three-wave bull impulsecycle and a three-wave bear connector cycle. My labeling conven-tion to identify the new four-wave cycle is Ai.

Because of the constraints imposed by the three logical opera-tors (�, �, and �) between adjacent waves, each three-wave cyclemay be merged only with three of the nine possible three-wavecycles. For example, the bull impulse cycle can be overlapped onlywith a bear connector, a bear expanding ascending triangle, or abear expanding symmetrical triangle. These are the three cyclesin which the height of the second wave is always greater than theheight of the first wave.

Multiple-Wave Cycles 105

1

7

2

3

6

4

5

Figure 18-3 Six-wave cycle.

TABLE 18-1 Component Waves

Components Cycle Name Cycle ID

1–2–3–4 Bull impulse A2–3–4–5 Bear connector i3–4–5–6 Bull contracting symmetrical triangle F4–5–6–7 Bear impulse a

To further clarify the naming convention for four-wave cycles,I present one more example (see Figure 18-4).

The three waves on the left in this figure form a bear contract-ing symmetrical triangle, whereas the three waves on the right forma bull impulse cycle. Thus the four-wave cycle in this figure is designated as eA.


Figure 18-4 Four-wave bear cycle.


Fourth Wave

OVERVIEW

Using the same logical height relationship that was applied to three-wave cycles, there are 27 possible four-wave bull patterns(3 � 3 � 3 � 27).

FREQUENCIES

Table 19-1 is a frequency table that is sorted in descending orderby 25-pip reversal amount.

Interesting to note in this table is the sharp decline in frequencybeginning with the cycle FI. All cycles above FI use only the � or �

relationship.

FOURTH-WAVE FORECAST

Given any three-wave cycle, we can estimate the relationshipbetween the third and fourth waves using Table 19-2 sorted by cyclename.

107


In the next example I will assume that a 15-pip swing reversalamount was used to create the swing data being scrutinized and that the last three-wave cycle was a bull connector cycle (seeFigure 19-1).

The three possible continuations for the overlapping cycle are acontracting symmetrical bear cycle (ID � e), a contracting descend-ing bear cycle (ID � d), and a bear connector cycle (ID � i)(see Figure 19-2).


TABLE 19-1 Four-Wave Cycles Sorted by 25-Pip Reversal

Cycle 5-Pip 15-Pip 25-Pip

IA 6.84 18.16 19.73AI 6.76 18.69 18.93HI 5.07 10.39 12.59AH 3.47 9.46 12.28IE 5.00 10.89 11.87EA 3.45 10.08 11.64HH 0.82 2.87 4.27EE 0.85 3.23 3.78FI 7.03 2.20 0.79DF 6.54 2.41 0.72ID 7.01 2.22 0.72AG 3.11 1.33 0.57GF 1.52 0.81 0.42CA 3.05 1.24 0.42FH 2.60 1.24 0.38ED 2.62 1.16 0.26GC 4.05 0.85 0.23DC 1.53 0.74 0.23HG 1.01 0.32 0.08CE 1.07 0.34 0.04DB 5.02 0.37 0.04BF 4.97 0.35 0.04GB 1.94 0.11 0.00BB 5.84 0.14 0.00BC 1.99 0.13 0.00CD 3.46 0.14 0.00FG 3.39 0.13 0.00

Forecasting the Fourth Wave 109

Figure 19-1 Three-wave bull connector cycle.

TABLE 19-2 Four-Wave Cycle Percentages Sorted by Cycle ID

Cycle 5-Pip 15-Pip 25-Pip

AG 3.11 1.33 0.57AH 3.47 9.46 12.28AI 6.76 18.69 18.93BB 5.84 0.14 0.00BC 1.99 0.13 0.00BF 4.97 0.35 0.04CA 3.05 1.24 0.42CD 3.46 0.14 0.00CE 1.07 0.34 0.04DB 5.02 0.37 0.04DC 1.53 0.74 0.23DF 6.54 2.41 0.72EA 3.45 10.08 11.64ED 2.62 1.16 0.26EE 0.85 3.23 3.78FG 3.39 0.13 0.00FH 2.60 1.24 0.38FI 7.03 2.20 0.79GB 1.94 0.11 0.00GC 4.05 0.85 0.23GF 1.52 0.81 0.42HG 1.01 0.32 0.08HH 0.82 2.87 4.27HI 5.0780 10.39 12.59IA 6.84 18.16 19.73ID 7.01 2.22 0.72IE 5.00 10.89 11.87

From Table 19-2. we find the frequencies listed in Table 19-3 forthe connector cycle.

Thus there is an over 65 percent likelihood (58.08 � 7.10) thatthe fourth wave will touch or descend below the lowest point inthe third wave when using a 15-pip reversal amount.


Figure 19-2 Possible fourth-wave continuations

TABLE 19-3 Connector Cycle Percentages

Cycle ID Raw Percent Adjusted Percent

IA 18.16 58.08ID 2.22 7.10IE 10.89 34.82


Fifth Wave

PRUNING

As the minimum reversal amount is increased, five of the ninebasic three-wave cycles decrease in frequency of occurrence owingto the equality relationship. I will arbitrarily refer to these cyclesas secondary cycles, indicating that their low frequency equates toless importance. Specifically, these cycles are listed in Table 20-1

The four remaining cycle types will be referred to as primarycycles because their greater frequency of occurrence implies greaterimportance. They are listed in Table 20-2.

The purpose for making this distinction is to lessen the numberof cycle permutations that will occur when examining cycles com-posed of more than four waves. Using the overlapping methodologyto name all five-wave cycles, there are 81 possibilities (9 � 3 � 3).

111

TABLE 20-1 Secondary Cycle Types

Cycle ID Cycle Name

B RectangleC Contracting descending triangleD Contracting ascending triangleF Expanding ascending triangleG Expanding descending triangle


However, by removing all the secondary cycles, this number isreduced significantly to only 16 permutations (4 � 2 � 2). In otherwords, I am dropping the equal to condition from the height-ratiocriteria. I am now concerned only with greater than or less thanheight ratios. This form of pruning is justified mathematicallybecause it keeps the data and the results more manageable.

FREQUENCIES

In the Table 20-3, cycles are sorted in descending order of thefourth column.

There is no surprise that the trending cycles IAI and AIA leadthe list. It is interesting to note, though, that the 5- and 15-pipreversal amounts do not exactly match the order of the 25-pipreversal amount.


TABLE 20-2 Primary Cycle Types

Cycle ID Cycle Name

A ImpulseE Contracting Symmetrical triangleH Expanding symmetrical triangleI Connector

TABLE 20-3 Five-Wave Cycles Sorted by 25-Pip Reversal Amount

Cycle-ID 5-Pips 15-Pips 25-Pips

IAI 15.92 14.91 13.14AIA 11.78 14.40 12.70AHI 10.92 8.97 9.46IEA 10.77 9.43 9.10HIA 6.84 7.01 7.84IAH 7.52 7.11 7.64AIE 6.53 7.59 7.12EAI 7.25 7.86 6.79EAH 4.12 4.17 5.38HIE 8.05 5.26 5.30HHI 2.91 3.01 3.72AHH 1.93 2.70 3.60IEE 2.86 3.22 3.40EEA 1.91 3.06 3.32HHH 0.30 0.55 0.85EEE 0.39 0.73 0.65

FIFTH-WAVE FORECAST

Table 20-4, sorted by cycle ID, provides the information necessaryto calculate probabilities for the binary relationships greater thanand less than for height ratios.

EXTENDED IMPULSE CYCLE

In my naming convention, I define a four-wave extended impulsecycle as Ai because it has the potential to become a five-wave Elliotimpulse cycle (see Figure 20-1).

Using a 15-pip reversal amount and referring to Table 20-1, wecan calculate the likelihood that a full five-wave bull Elliott impulsecycle (AiA) will come to fruition:

AiA � 14.40AiE � 7.59Probability of AiA � 100 � 14.40/(14.40 � 7.59)

� 65.5 percent

Forecasting the Fifth Wave 113

TABLE 20-4 Five-Wave Cycle Frequencies Sorted by Cycle ID

Cycle-ID 5-Pips 15-Pips 25-Pips

AHH 1.93 2.70 3.60AHI 10.92 8.97 9.46AIA 11.78 14.40 12.70AIE 6.53 7.59 7.12EAH 4.12 4.17 5.38EAI 7.25 7.86 6.79EEA 1.91 3.06 3.32EEE 0.39 0.73 0.65HHH 0.30 0.55 0.85HHI 2.91 3.01 3.72HIA 6.84 7.01 7.84HIE 8.05 5.26 5.30IAH 7.52 7.11 7.64IAI 15.92 14.91 13.14IEA 10.77 9.43 9.10IEE 2.86 3.22 3.40

This 66 percent likelihood gives some credibility to the ancientmaxim, “A trend continues until it ends.”

POTENTIAL HEAD AND SHOULDERS

Head-and-shoulders patterns represent trend reversals and, assuch, should be examined closely. The head-and-shoulders patternthat I will analyze has the critical descending neckline (ID � Ah)(see Figure 20-2).

The first indication that the highest point in the third wave isin fact a major reversal point is that the fifth wave does not exceedthis point. Thus we need to calculate the probability that the cycle


Figure 20-1 Potential five-wave Elliott impulse cycle,

Figure 20-2 Head and shoulders with descending neckline.

AhI will occur. Again, refer to Table 20-1, this time using a 5-pipreversal amount:

AhI � 10.92AhH � 1.93Probability of AhI � 100 � 10.92/(10.92 � 1.93)

� 85.0 percent

This is a very promising result. There is an 85 percent likelihoodthat the fifth wave will be the beginning of a downward trend. Notethe relation between this pattern and the preceding one, theextended impulse cycle, where the height of the fourth wave is theonly difference. The extended impulse cycle has the potential tobe a head and shoulders pattern with an ascending neckline, butthe probability is far less.

EXTENDED CONTRACTING TRIANGLE

From classic pattern-recognition theory, recall that triangles normally are considered to be continuation harbingers. This maybe an optimistic appraisal that you feel requires testing. Let’sexamine the bull contracting symmetrical triangle (ID � Ee)using a 25-pip reversal amount (see Figure 20-3).

My premise is that the trend of the first wave is the current over-all trend. Therefore, I want to calculate the likelihood that theheight of the fifth wave will be greater than the height of thefourth wave:


Figure 20-3 Extended contracting symmetrical triangle.

EeA � 3.32EeE � 0.65Probability of EeA � 100 � 3.32/ (3.32 � 0.65)

� 83.6 percent

This implies that there is an 84 percent likelihood that the exist-ing trend prior to a double contracting symmetrical triangle willcontinue. Contracting triangles possess a unique fascination, notunlike increasing the torque on a metal spring to a critical point.Sharp breakouts frequently follow these triangles.

EXTENDED EXPANDING TRIANGLE

Extended expanding triangles are a sign that the market is bothhighly volatile and very confused. Perhaps indecisive is a betterword (see Figure 20-4).

Expanding triangles are the only pattern that cannot be main-tained indefinitely. In fact, in the 29 million� EURUSD databasecovering the tick data for calendar year 2005, there was only oneoccurrence of an HhHhH cycle, a seven-wave expanding triangle.Therefore, I will calculate the probability that the height of thefifth wave in this cycle will be less than the height of the fourthwave, thus terminating the expanding phenomenon. I will employa 5-pip reversal amount in this example:

HhI � 2.91HhH � 0.30


Figure 20-4 Extended expanding symmetrical triangle.

Probability of HhI � 100 � 2.91/(2.91 � 0.30)� 90.7 percent

This result is not surprising, merely good to know, particularlywhen dealing with higher swing reversal amounts. I also shouldnote that contracting triangles theoretically can be maintainedover time simply by entering a phase of zero activity, where the lastprice lingers indefinitely (such as during weekends).

This 91 percent probability represents about as close to a “surething” as is possible when forecasting with wave theory.



Sixth Wave

OVERVIEW

Using the three logical operators �, �, and � for height-ratio comparisons of six-wave cycles generates 243 permutations (9 � 3� 3 � 3). By excluding the equality relationship, the number ofpermutations is reduced to 32 possibilities (4 � 2 � 2 � 2).

FREQUENCIES

Table 21-1 is a six-wave cycle frequency table that is sorted indescending order on the fourth column.

Note the visible decrease in frequency between the second andthird rows. Also again we find the extended contracting andexpanding triangles at the bottom of the frequency list.

When analyzing cycles composed of six waves, we must note that these are not just randomly connected cycles. My overlappingwave scheme ensures that the four central three-wave cycles aretightly linked to the adjacent cycles using the logical height-ratiocomparisons.

119


Table 21-2 is a table of six-wave cycles that has been sorted bycycle ID to facilitate forecasting the sixth wave.

EXTENDED IMPULSE CYCLE

Given the five-wave impulse cycle shown in Figure 21-1, I want todetermine the likelihood that the sixth wave will be shorter thanthe fifth wave, thereby facilitating still another bull wave continu-ing the upward trend.


TABLE 21-1 Six-Wave Cycles Sorted by 25-Pip Reversal Amount

Cycle 5-Pips 15-Pips 25-Pips

AIAI 8.83 9.70 9.26IAIA 9.17 9.33 9.07IEAI 7.50 6.33 6.02IAHI 7.67 5.73 5.71AIEA 4.85 5.66 5.55IAIE 4.91 5.36 5.27EAIA 3.90 5.40 5.20AHIA 4.82 4.94 5.11HIAI 5.24 5.03 5.07AIAH 4.15 4.94 4.95AHIE 5.47 3.92 3.85HIEA 5.29 3.66 3.71IEAH 3.93 3.18 3.40EAHI 4.06 3.22 3.36IEEA 2.06 2.48 2.54AHHI 2.19 2.34 2.47EAIE 2.34 2.38 2.35HIAH 2.42 2.08 2.21EEAI 1.18 1.87 1.90HHIA 1.16 1.75 1.85AIEE 1.00 1.74 1.79IAHH 1.35 1.61 1.71HIEE 1.80 1.44 1.43HHIE 1.69 1.21 1.30EAHH 0.76 1.11 1.22EEAH 0.83 1.13 1.16IEEE 0.43 0.69 0.68EEEA 0.30 0.65 0.64HHHI 0.36 0.53 0.59AHHH 0.24 0.47 0.53HHHH 0.03 0.06 0.06EEEE 0.05 0.06 0.05

The frequencies from Table 21-1 using a 15-pip reversal amountare

AiAi � 9.70AiAh � 4.94Probability of AiAi � 100 � 9.70/(9.70 � 4.94)

� 66.3 percent

Forecasting the Sixth Wave 121

TABLE 21-2 Six-Wave Cycles Sorted by Cycle ID

Cycle 5-Pips 15-Pips 25-Pips

AHHH 0.24 0.47 0.53AHHI 2.19 2.34 2.47AHIA 4.82 4.94 5.11AHIE 5.47 3.92 3.85AIAH 4.15 4.94 4.95AIAI 8.83 9.70 9.26AIEA 4.85 5.66 5.55AIEE 1.00 1.74 1.79EAHH 0.76 1.11 1.22EAHI 4.06 3.22 3.36EAIA 3.90 5.40 5.20EAIE 2.34 2.38 2.35EEAH 0.83 1.13 1.16EEAI 1.18 1.87 1.90EEEA 0.30 0.65 0.64EEEE 0.05 0.06 0.05HHHH 0.03 0.06 0.06HHHI 0.36 0.53 0.59HHIA 1.16 1.75 1.85HHIE 1.69 1.21 1.30HIAH 2.42 2.08 2.21HIAI 5.24 5.03 5.07HIEA 5.29 3.66 3.71HIEE 1.80 1.44 1.43IAHH 1.35 1.61 1.71IAHI 7.67 5.73 5.71IAIA 9.17 9.33 9.07IAIE 4.91 5.36 5.27IEAH 3.93 3.18 3.40IEAI 7.50 6.33 6.02IEEA 2.06 2.48 2.54IEEE 0.43 0.69 0.68

Thus there is a two out of three chance that a five-wave impulsecycle will continue trending in the same direction. This formationshould be confirmed by using different swing reversal amounts.

HEAD AND SHOULDERS

A five-wave head and shoulders formation with a descendingneckline is designated AhI. I want to calculate the probability thatthe sixth wave will fall below the vertex of the fourth and fifthwaves, thus confirming that a reversal is actually in progress (seeFigure 21-2).


Figure 21-1 Extended impulse cycle.

Figure 21-2 Head and shoulders forecast.

Using a 25-pip reversal amount, I obtain the following fromTable 21-1:

AhIa � 5.11AhIe � 3.85Probability of AhIa � 100 � 5.11/(5.11 � 3.85)

� 57.0 percent

I also will calculate the probability using a 5-pip reversal amount:

AhIa � 4.82AhJe � 5.47Probability of AhIa � 100 � 4.82/(4.82 � 5.47)

� 46.8 percent

Fortunately, this inconsistency, where probabilities lie on bothsides of the 50 percent median when using different reversalamounts, is very rare. When in doubt, stay out of the market.

CHIMERA CYCLE

In this example I will analyze a five-wave chimera cycle so namedbecause of its changeling properties. It starts as a contracting sym-metrical triangle and ends as an expanding symmetrical triangle(ID � EaH). This cycle occurs frequently during lateral congestion(see Figure 21-3).

Forecasting the Sixth Wave 123

Figure 21-3 Chimera cycle forecast.

I want to test to determine if the sixth wave will drop below thevertex of the fourth and fifth waves. This time I will use a 15-pipreversal amount in Table 21-1:

EaHi � 3.22EaHh � 1.11Probability of EaHi � 100 � 3.22/(3.22 � 1.11)

� 74.4 percent

There is a 74 percent likelihood that the sixth wave will notdescend below the vertex of the fourth and fifth waves.


Chapter 22Advanced

Features

FRACTAL LEVELS

In earlier chapters I mentioned using swing analysis at differentfractal levels to determine the possible strength of a subsequentwave and its component waves. One method to perform this oper-ation is to calculate swing data using two different minimum rever-sal amounts, say, a 5-pip reversal amount for the child fractal dataand a 10-pip reversal amount for the parent swing data.

To perform this operation successfully, both sets of swing datamust end on the same point, preferably the last known price in theraw data. Then, by using height-ratio wave comparisons at two levels, the trader can determine if the predicted waves will extendin the same price direction.

The disadvantage of this method is that prolonged experimen-tation may be required to find a “perfect” child-parent fractal rela-tionship. As markets become more volatile, higher reversal amountsmay be required to filter out undesired white noise. As markets losevolatility, lower reversal amounts may be preferable. Also, there isthe possibility that the difference in the two reversal amounts maybe set too high. This will cause the fractal relationship to jump froma child-parent relationship to a child-grandparent relationship. An

125


alternative method that progresses through each fractal level oneat a time is given below.

PIVOT CHARTS

The pivot chart is a subset of the swing reversal chart that uses thebare minimum criteria as the parameters of the reversal algorithm.That is, the reversal increment is set to the minimum legal pricefluctuation in the time series, and the number of reversal units isset to one. Paradoxically, this chart does not filter any reversals inprice. The reason for this will become apparent shortly.

The pivot chart does, however, have the advantage of merginga multipoint trend into a single diagonal line. It also filters out horizontal price movements. Another advantage is that a pivotchart can be constructed easily by hand without the use of a soft-ware utility. All continuous upward movements resolve to a dia-gonal line, as do all continuous downward movements. There isno need to calculate a minimum reversal amount with each newprice. The pivot chart is the only swing chart that can never havea child fractal chart. In other words, it is the lowest commondenominator of all swing charts (see Figure 22-1).


Figure 22-1 Pivot chart.

I note in the figure that no reversal is filtered out. A standardswing reversal chart of the same raw data (the solid curve) usinga 15-pip reversal amount would generate only two diagonals, onebull wave followed by one bear wave.

CHANNEL LINES

The classification of the basic three-wave cycles discussed earlierwill employ a rigid methodology that prevents any ambiguity—channeling. This is a purely mathematical process that has nothingto do with excursions into the supernatural.

All three-wave cycles always can be inscribed inside a channelquadrangle or triangle using the coordinate system in Table 22-1(see Figure 22-2).

Advanced Features 127

TABLE 22-1 Cycle Coordinate System

Wave No. Starting Point Ending Point

1 x1,y1 x2,y2

2 x2,y2 x3,y3

3 x3,y3 x4,y4

x1,y1

x2,y2

x3,y3

x4,y4

Figure 22-2 Impulse cycle with coordinates.

The upper boundary of the channel polygon (either a triangleor a parallelogram) passes through the two points x2, y2 and x4, y4,whereas the lower boundary of the channel shape passes throughthe two points x1, y1 and x3, y3, as depicted in Figure 22-3.

The criteria for classifying any three-wave cycle are derived fromthe slopes of its two identifying channel lines. In Figure 22-2, bothslopes are positive, or travel from the lower left to the upper rightof the diagram (the standard bull impulse cycle described earlier).The slope property of either channel line may have one of threemathematical attributes: positive, negative, or zero. This equates toupward, downward, and horizontal price movements.

The nine basic bull cycles with channel lines are shown in Figure 22-4.

Concerning the naming convention for triangles, the modifiercontracting signifies that the channel lines converge on the rightside of the cycle. Conversely, expanding indicates that the channellines converge on the left of the cycle.

CYCLE COLLAPSING

The pivot chart described earlier facilitates a very simple and directapproach to determining the parent fractal level of any swing data,which I refer to as cycle collapsing. To collapse any three-wave bullcycle shown in Figure 22-4 into a single wave, the following tworules must be observed:


Figure 22-3 Three-wave impulse cycle with channel lines.

1. Using the coordinate system in Table 22-1, the starting point y1must be lower than or equal to point y3.

2. The ending point y4 must be greater than or equal to point y2.

If both conditions are satisfied, then the collapsing method ispurely mechanical: Extend a straight line from the cycle startingcoordinates x1, y1 to the cycle end point x4, y4. Examples areshown in Figure 22-5.

The four examples in this figure are, in fact, the only bull cyclesthat can be collapsed into a single wave in one operation. Theremaining five bull cycles (contracting descending triangle, con-tracting symmetrical triangle, expanding descending triangle,expanding symmetrical triangle, and the connector cycle) all

Advanced Features 129

(A) Impulse

(D) Contracting DescendingTriangle

(E) Contracting SymmetricalTriangle

(G) Expanding DescendingTriangle

(H) Expanding SymmetricalTriangle

(B) Rectangle (C) Contracting AscendingTriangle

(F) Expanding AscendingTriangle

(I) Connector

Figure 22-4 The nine basic bull cycles with channel lines.

violate one of the preceding rules (the bull connector cycle vio-lates both rules).

Surprisingly, though, a second pass through the collapsed dataagain will collapse several three-wave cycles into a single wave. Nomore than two passes through the data are required to convert aset of swing data to their parent fractal data.


Bull Impulse

Contracting Ascending Triangle

Bull Rectangle

Expanding Ascending Triangle

Figure 22-5 Collapsing bull cycles.

PART 5Shockwave Raw

Data


Chapter 23OHLC Bar Charts

OVERVIEW

In this chapter I will begin my analysis of shockwaves, with the visualrepresentation of the data displayed as 1-minute OHLC quotes. Toavoid any bias in the selection of the date ranges involved, I will usethe first 15 Fridays in 2006 as my database. I will focus on 8:30 a.m.ET because this is when most of the scheduled news announce-ments are released.

In the charts that follow, the time frame is maintained uniformfor all dates: 8:00 to 10:00 a.m. There are three distinct phases during this time frame:

1. The preshockwave calm from 8:00 to 8:29 a.m.

2. The shockwave surge from 8:30 a.m. until a significant extremeis reached.

3. The reactive response immediately following the significant extreme.

At the top of each chart are the 1-minute OHLC quotes repre-sented as vertical bars. The horizontal background grid is scaled inincrements of a single pip. Each OHLC region has two dotted lines.The vertical dotted line indicates the 8:30 longitude, whereas thehorizontal dotted line is aligned with the opening price in the 8:30quote. The vertical numeric scale on the right also represents pips.

133


In the middle of each chart is the activity oscillator with upticksrepresented as light rectangles and downticks as shaded rectangles.Their sum is displayed in the right vertical scale as total ticks perminute.

At the bottom of each chart is the 1-minute range oscillator. Thelight rectangle at the top represents the distance between the openand high prices. The shaded rectangle below it represents the dis-tance from the open to the closing price. Range is measured inpips on the right vertical scale.

Beneath each chart are the descriptive statistic values coveringthe 2-hour period for that chart.

STUDY NO. 1: JANUARY 6, 2006

134 Part 5: Shockwave Raw Data

1.21781.21731.21681.21631.21581.21531.21481.21431.21381.21331.21281.21231.21181.21131.21081.21031.20981.20931.2088

525450375300225150750

42363024181260

10000930090008300800

EURUSD 1/6/2006 Fri 0800 to 1000 ET 1-Min OHLC, 1-Min Activity and 1-Min Range


OHLC Bar Charts 135

1.2086

1.2081

1.2076

1.2071

1.2066

1.2061

1.2056

1.2051

1.2046

1.2041

1.2036

52545037530022515075

0

15129630

10000930090008300800


Figure 23-1 OHLC, activity and range, January 6, 2006.

Open 1.2101

High 1.2182

Low 1.2087

Close 1.2154

Midrange 1.2135

Range 0.0095

Mean 1.2133

Standard Deviation 0.0026




Open 1.2042

High 1.2090

Low 1.2035

Close 1.2052

Midrange 1.2063

Range 0.0055

Mean 1.2058

Standard deviation 0.0012

1.2116

1.2111

1.2106

1.2101

1.2096

1.2091

1.2086

1.2081

1.2076

1.2071

1.2066

30025020015010050

0

1086420


0800 0830 0900 0930 1000


OHLC Bar Charts 137


Open 1.2075

High 1.2117

Low 1.2065

Close 1.2094

Midrange 1.2091

Range 0.0052

Mean 1.2087


1.22361.22311.22261.22211.22161.22111.22061.22011.21961.21911.21861.21811.21761.21711.21661.2161

525450375300225150750

3025201510

50

10000930090008300800


STUDY NO. 5: FEBRUARY 3, 2006



Open 1.2183

High 1.2238

Low 1.2160

Close 1.2222

Midrange 1.2199

Range 0.0078

Mean 1.2208


EURUSD 2/3/2006 Fri 0800 to 1000 ET1-Min OHLC, 1-Min Activity and 1-Min Range

1.21011.20941.20871.20801.20731.20661.20591.20521.20451.20381.20311.20241.20171.20101.20031.19961.19891.19821.19751.1968

60052545037530022515075

040353025201510

50

0800 0830 0900 0930 1000


OHLC Bar Charts 139

Figure 23-5 OHLC, activity and range, February 3, 2006.

Open 1.2064

High 1.2105

Low 1.1967

Close 1.1982

Midrange 1.2036

Range 0.0138

Mean 1.2018


1.2028

1.2023

1.2018

1.2013

1.2008

1.2003

1.1998

1.1993

1.1988

1.1983

1.1978

1.1973

1.1968

1.1963

525450375300225150750

129630

10000930090008300800





Open 1.1974

High 1.2028

Low 1.1962

Close 1.2015

Midrange 1.1995

Range 0.0066

Mean 1.1998


1.1897

1.1894

1.1891

1.1888

1.1885

1.1882

1.1879

1.1876

1.1873

1.1870

1.1867

1.1864

1.1861

1.1858

375

300

225

150

75

0

1086420

0800 0830 0900 0930 1000



OHLC Bar Charts 141


Open 1.1866

High 1.1899

Low 1.1857

Close 1.1897

Midrange 1.1878

Range 0.0042

Mean 1.1875


1.1915

1.1912

1.1909

1.1906

1.1903

1.1900

1.1897

1.1894

1.1891

1.1888

45037530022515075

0

121086420

0800 0830 0900 0930 1000


STUDY NO. 9: MARCH 3, 2006



Open 1.1895

High 1.1916

Low 1.1887

Close 1.1899

Midrange 1.1902

Range 0.0029

Mean 1.1897


1.2055

1.2052

1.20491.20461.20431.2040

1.2037

1.2034

1.2031

1.2028

1.2025

1.20221.2019

1.2016

300250200150100500

8

6

4

2

0

0800 0830 0900 0930 1000



OHLC Bar Charts 143

Figure 23-9 OHLC, activity and range, March 3, 2006.

Open 1.2023

High 1.2055

Low 1.2015

Close 1.2050

Midrange 1.2035

Range 0.0040

Mean 1.2036


1.19231.19201.19171.19141.19111.19081.19051.19021.18991.18961.18931.18901.18871.18841.18811.18781.18751.18721.18691.18661.18631.1860

450375300225150750

24201612840

0800 0830 0900 0930 1000





Open 1.1921

High 1.1925

Low 1.1859

Close 1.1873

Midrange 1.1892

Range 0.0066

Mean 1.1895



0800 0830 0900 0930 1000

1.21991.21961.21931.21901.21871.21841.21811.21781.21751.21721.21691.21661.21631.21601.21571.21541.2151

35030025020015010050

0

12

9

6

3

0


OHLC Bar Charts 145


Open 1.2186

High 1.2201

Low 1.2150

Close 1.2155

Midrange 1.2176

Range 0.0051

Mean 1.2181


0800 0830 0900 0930 1000

1.1988

1.1985

1.1982

1.1979

1.1976

1.1973

1.1970

1.1967

1.1964

1.1961

1.1958

1.1955

1.1952

450375300225150750

12

9

6

3

0





Open 1.1969

High 1.1990

Low 1.1951

Close 1.1972

Midrange 1.1971

Range 0.0039

Mean 1.1971



1.21341.21311.21281.21251.21221.21191.21161.21131.21101.21071.21041.21011.20981.20951.20921.20891.20861.2083

525450375300225150

750

6

15129

30

0800 0830 0900 0930 1000

STUDY NO. 14: APRIL 7, 2006

OHLC Bar Charts 147


Open 1.2103

High 1.2136

Low 1.2082

Close 1.2103

Midrange 1.2109

Range 0.0054

Mean 1.2106


1.22261.22211.22161.22111.22061.22011.21961.21911.21861.21811.21761.21711.21661.21611.21561.21511.2146

600525450375300225150750

2824201612840

0800 0830 0900 0930 1000




Figure 23-14 OHLC, activity and range, April 7, 2006.

Open 1.2179

High 1.2227

Low 1.2145

Close 1.2157

Midrange 1.2186

Range 0.0082

Mean 1.2181



1.21161.21151.21141.21131.21121.21111.21101.21091.21081.21071.21061.21051.21041.21031.21021.2101

907560453015086420

0800 0830 0900 0930 1000

SUMMARY OF STATISTICS

See Table 23-1. The open, high, low, and close are displayed interms of the quote currency USD. Range and standard deviationare expressed as pips (0.0001 USD).

OHLC Bar Charts 149

Figure 23-15 OHLC, activity and range, April 14, 2006.

Good Friday

Open 1.2109

High 1.2116

Low 1.2101

Close 1.2105

Midrange 1.2109

Range 0.0015

Mean 1.2110


TABLE 23-1 Summary of Statistics

Study Date Open High Low Close Range Std Dev

1 1/6/2006 1.2101 1.2182 1.2087 1.2154 95 26

2 1/13/2006 1.2042 1.2090 1.2035 1.2052 55 12

3 1/20/2006 1.2075 1.2117 1.2065 1.2094 52 13

4 1/27/2006 1.2183 1.2238 1.2160 1.2222 78 21

5 2/3/2006 1.2064 1.2105 1.1967 1.1982 138 34

6 2/10/2006 1.1974 1.2028 1.1962 1.2015 66 20

7 2/17/2006 1.1866 1.1899 1.1857 1.1897 42 9

8 2/24/2006 1.1895 1.1916 1.1887 1.1899 29 4

9 3/3/2006 1.2023 1.2055 1.2015 1.2050 40 9

10 3/10/2006 1.1921 1.1925 1.1859 1.1873 66 19

11 3/17/2006 1.2186 1.2201 1.2150 1.2155 51 9

12 3/24/2006 1.1969 1.1990 1.1951 1.1972 39 5

13 3/31/2006 1.2103 1.2136 1.2082 1.2103 54 8

14 4/7/2006 1.2179 1.2227 1.2145 1.2157 82 20

15 4/14/2006 1.2109 1.2116 1.2101 1.2105 15 2

RANGE RATIO

When analyzing interval data as opposed to streaming tick data,an interesting phenomenon occurs. Recall that I define time frameas the overall duration of the chart data from start date to enddate. The time interval is defined as the equal-sized gradationswithin the time frame. Thus, in the preceding studies, the timeframe is from 8:00 to 10:00 a.m., and the time interval is 1 minute,which amounts to 121 same-sized gradations or number of inter-vals per frame.

The range statistic for each study covers the entire time framefor that one study. This gives rise to a new volatility gauge basedon the average interval range compared with the range of the totaltime frame (see Table 23-2).

The calculated columns are defined as follows:

Average range is calculated as the sum of individual interval rangesdivided by the number of intervals N (121).

Total range is the range of the entire time frame.Total range/N is the total range divided by the number of intervals.


TABLE 23-2 Range Ratio Index

Study Date Average Range Total Range Total Range/N Range Ratio

1 1/6/2006 9.66 95 0.7851 8.132 1/13/2006 6.60 55 0.4545 6.893 1/20/2006 6.14 52 0.4298 7.004 1/27/2006 7.17 78 0.6446 9.995 2/3/2006 8.76 138 1.1405 13.026 2/10/2006 7.18 66 0.5455 7.597 2/17/2006 6.05 42 0.3471 5.748 2/24/2006 5.60 29 0.2397 4.289 3/3/2006 5.31 40 0.3306 6.23

10 3/10/2006 8.21 66 0.5455 6.6411 3/17/2006 6.44 51 0.4215 6.5512 3/24/2006 5.99 39 0.3223 5.3813 3/31/2006 6.76 54 0.4463 6.6014 4/7/2006 8.83 82 0.6777 7.6815 4/14/2006 4.31 15 0.1240 2.87

Range ratio is 100 times total range divided by the average range.

The range ratio index (a single value for the entire sample) canbe converted easily to a moving indicator with a lookback periodcalled the range ratio oscillator.

Range ratio oscillatorx � 100 � total rangex___________________________(Σ average rangex – lag to x/N)

where x � the time index in the interval arraylag � the number of time units in the lookback period

The purpose of the range ratio oscillator is to determine if amarket is volatile enough to warrant a market entry order. If therange of the individual intervals is too high in proportion to therange of the lookback period, then this is an indicator that lateralwhipsawing may be in progress. For scalpers who traditionally setnarrow protective orders, this means that stop-loss limit orders maybe triggered very unexpectedly. When the range of the individualintervals is sufficiently low in relation to the range of the lookbackperiod, then stop-loss orders will be triggered only when a well-defined reversal trend has begun.

OBSERVATIONS

Economic news announcements are almost never scheduled forbanking holidays, which greatly reduces the volatility and liquid-ity of the currency markets, even though the markets remainopen. Under normal conditions, low-volatility trading periodsshould be avoided unless there is a perceived advantage to long-term trading (say, over 24 hours).

In the preceding studies, only no. 15 coincided with a holiday(Good Friday). I included study no. 15 simply to illustrate the point that diminished range rarely offers sufficient profits for the risk involved (particularly the automatic loss of the trans-action cost).

OHLC Bar Charts 151

By excluding the results of study no. 15, I find that in theremaining 14 studies the shockwave breakouts were in an upwarddirection. In study no. 7 and study no. 10, the shockwave was down-ward. Since the USD is the second currency in the EURUSD currency pair, this means that public opinion interpreted thenews releases as pessimistic 85.7 percent of the time.


Chapter 24Tick Charts

OVERVIEW

In Chapter 23, the 1-minute OHLC bar charts provide traders witha high-altitude aerial perspective of the shockwave landscape. Inthis chapter I have intentionally zoomed in to focus on the first 3minutes of the shockwave phenomenon.

Prices are plotted as a univariate continuous line because thereare no OHLC quotes in tick data. The horizontal lines in eachchart grid are graduated at 1-pip increments, with the first pricein the shockwave identified as a dotted horizontal line. The verti-cal gradations above the time labels at the bottom of each chartare every 25 ticks.

Descriptive statistics of the 3-minute data follow each chart. Sizerepresents the sum of upticks and downticks during the 3-minuteperiod.

153




1.2141

1.2138

1.2135

1.2132

1.2129

1.2126

1.2123

1.2120

1.2117

1.2114

1.2111

1.2108

1.2105

1.2102

1.2099

1.2096

0830 0831 0832

EURUSD 1/6/2006 Fri 0830 to 0831 ET Continuous Tick Chart

Figure 24-1 Tick chart, January 6, 2006.

Size 1,000

Open 1.2094

High 1.2142

Low 1.2094

Close 1.2125

Midrange 1.2118

Range 0.0048

Mean 1.2128



Tick Charts 155


Size 781

Open 1.2052

High 1.2068

Low 1.2039

Close 1.2065

Range 0.0029

Mean 1.2053


EURUSD 1/13/2006 Fri 0830 to 0832 ETContinuous Tick Chart

1.2067

1.2065

1.2063

1.2061

1.2059

1.2057

1.2055

1.2053

1.2051

1.2049

1.2047

1.2045

1.2043

1.2041

1.2039

0830 0831 0832




1.2072

1.2071

1.2070

1.2069

1.2068

0830 0831 0832


Size 168

Open 1.2069

High 1.2072

Low 1.2068

Close 1.2069

Midrange 1.2070

Range 0.0004

Mean 1.2070



Tick Charts 157


Size 1,112

Open 1.2171

High 1.2223

Low 1.2170

Close 1.2214

Midrange 1.2197

Range 0.0053

Mean 1.2205


1.22221.22201.22181.22161.22141.22121.22101.22081.22061.22041.22021.22001.21981.21961.21941.21921.21901.21881.21861.21841.21821.21801.21781.21761.21741.21721.2170

0830 0831 0832





1.21041.21021.21001.20981.20961.20941.20921.20901.20881.20861.20841.20821.20801.20781.20761.20741.20721.20701.20681.20661.20641.20621.20601.20581.20561.20541.20521.20501.20481.20461.20441.20421.2040

0830 0831 0832

Figure 24-5 Tick chart, February 3, 2006.

Size 1,111

Open 1.2062

High 1.2105

Low 1.2040

Close 1.2046

Midrange 1.2073

Range 0.0065

Mean 1.2064



Tick Charts 159


Size 849

Open 1.1969

High 1.1978

Low 1.1962

Close 1.1975

Midrange 1.1970

Range 0.0016

Mean 1.1972


1.1978

1.1977

1.1976

1.1975

1.1974

1.1973

1.1972

1.1971

1.1970

1.1969

1.1968

1.1967

1.1966

1.1965

1.1964

1.1963

1.1962

0830 0831 0832




1.1869

1.1868

1.1867

1.1866

1.1865

1.1864

1.1863

1.1862

1.1861

1.1860

1.1859

1.1858

0830 0831 0832



Size 592

Open 1.1869

High 1.1869

Low 1.1858

Close 1.1861

Midrange 1.1864

Range 0.0011

Mean 1.1864



Tick Charts 161


1.1912

1.1911

1.1910

1.1909

1.1908

1.1907

1.1906

1.1905

1.1904

1.1903

1.1902

1.1901

1.1900

1.1899

1.1898

1.1897

1.1896

0830 0831 0832


Size 741

Open 1.1898

High 1.1912

Low 1.1896

Close 1.1905

Midrange 1.1904

Range 0.0016

Mean 1.1904




1.2034

1.2033

1.2032

1.2031

1.2030

1.2029

1.2028

1.2027

1.2026

1.2025

1.2024

1.2023

0830 0831 0832


Figure 24-9 Tick chart, March 3, 2006.

Size 268

Open 1.2028

High 1.2034

Low 1.2023

Close 1.2030

Midrange 1.2029

Range 0.0011

Mean 1.2029



Tick Charts 163


1.19251.19241.19231.19221.19211.19201.19191.19181.19171.19161.19151.19141.19131.19121.19111.19101.19091.19081.19071.19061.19051.19041.19031.19021.19011.19001.18991.18981.18971.18961.18951.1894

0830 0831 0832


Size 999

Open 1.1921

High 1.1925

Low 1.1894

Close 1.1905

Midrange 1.1910

Range 0.0031

Mean 1.1907





1.2190

1.2189

1.2188

1.2187

1.2186

1.2185

1.2184

0830 0831 0832


Size 308

Open 1.2187

High 1.2190

Low 1.2184

Close 1.2185

Midrange 1.2187

Range 0.0006

Mean 1.2186



Tick Charts 165


1.1980

1.1979

1.1978

1.1977

1.1976

1.1975

1.1974

1.1973

1.1972

1.1971

1.1970

1.1969

1.1968

1.1967

1.1966

1.1965

1.1964

1.1963

1.1962

1.1961

0830 0831 0832


Size 868

Open 1.1963

High 1.1980

Low 1.1961

Close 1.1972

Midrange 1.1971

Range 0.0019

Mean 1.1973





1.2117

1.2116

1.2115

1.2114

1.2113

1.2112

1.2111

1.2110

1.2109

0830 0831 0832


Size 204

Open 1.2113

High 1.2118

Low 1.2109

Close 1.2115

Midrange 1.2114

Range 0.0009

Mean 1.2114



Tick Charts 167


1.22271.22251.22231.22211.22191.22171.22151.22131.22111.22091.22071.22051.22031.22011.21991.21971.21951.21931.21911.21891.21871.21851.21831.21811.21791.21771.21751.21731.2171

0830 0831 0832

Figure 24-14 Tick chart, April 7, 2006.

Size 1,316

Open 1.2190

High 1.2227

Low 1.2171

Close 1.2219

Midrange 1.2199

Range 0.0056

Mean 1.2203





1.2110

1.2109

1.2108

1.2107

0830 0831 0832

Figure 24-15 Tick chart, April 14, 2006.

Size 56

Open 1.2111

High 1.2111

Low 1.2107

Close 1.2108

Midrange 1.2109

Range 0.0004

Mean 1.2109



See Table 24-1. Size represents the number of upticks anddownticks during the critical 3-minute shockwave period.

OBSERVATIONS

If the prevailing price does not move a minimum of 15 pips or morein the EURUSD pair between 8:30 and 8:32 a.m., then it is likelythat some anomaly has occurred. First, any or even all of theannouncements may have been postponed or canceled withoutadvance warning. Another possibility is that a bizarre state of trad-ing equilibrium has occurred, where orders from buyers and sell-ers are evenly matched. This can be determined if there is a sharpincrease in both upticks and downticks during the 3-minute period.

Using the arbitrary 15-pip divergence constraint on the 15 studies, 6 studies were upward, 2 studies were downward, and theremaining 7 studies failed to satisfy this condition during the first3 minutes. Nonetheless, the ratio of advances to declines is 3:1.

Tick Charts 169


Study Date Size Open High Low Close Range Std Dev

1 1/6/2006 1,000 1.2094 1.2142 1.2094 1.2125 48 62 1/13/2006 781 1.2052 1.2068 1.2039 1.2065 29 93 1/20/2006 168 1.2069 1.2072 1.2068 1.2069 4 14 1/27/2006 1,112 1.2171 1.2223 1.2170 1.2214 53 85 2/3/2006 1,111 1.2062 1.2105 1.2040 1.2046 65 136 2/10/2006 849 1.1969 1.1978 1.1962 1.1975 16 47 2/17/2006 592 1.1869 1.1869 1.1858 1.1861 11 28 2/24/2006 741 1.1898 1.1912 1.1896 1.1905 16 39 3/3/2006 268 1.2028 1.2034 1.2023 1.2030 11 2

10 3/10/2006 999 1.1921 1.1925 1.1894 1.1905 31 611 3/17/2006 308 1.2187 1.2190 1.2184 1.2185 6 112 3/24/2006 868 1.1963 1.1980 1.1961 1.1972 19 413 3/31/2006 204 1.2113 1.2118 1.2109 1.2115 9 114 4/7/2006 1,316 1.2190 1.2227 1.2171 1.2219 56 1615 4/14/2006 56 1.2111 1.2111 1.2107 1.2108 4 1

Another peculiarity that can plague scalpers who usually keepnarrow stop-loss orders is a phenomenon that I have named icicles. These occur when a distinct trend has developed, and sharpsingle breakouts exhibit themselves in the opposite direction.Icicles are the nemesis of traders who employ the trailing-stopstechnique, in which traders continually move stop-loss orders inthe direction of the trend. Examples of profit-depleting icicles canbe seen in studies 6, 10, and 14.


PART 6ShockwaveSwing Data


Chapter 25Swing Charts

OVERVIEW

In the preceding two chapters, shockwave data were presented intheir most pristine form: OHLC and tick charts. Readers are mostlikely intrigued with the diversity of form by which a shockwave canexhibit itself along both the x axis and the y axis.

In this chapter I have applied wave theory to the shockwave databetween 8:30 and 8:59 a.m. ET. This is not an arbitrarily selectedtime frame. In the studies where a decisive shockwave and its reac-tive phase are visibly discernible, the phenomena occurred duringthis period.

The standard swing reversal algorithm described in Chapter 15was applied to all 15 studies. Either a 10- or 15-pip reversal amountwas employed with only one exception (study no. 5 uses an 18-pipreversal amount). The resulting Forex wave theory nomenclatureis displayed inside square brackets in the chart headed “Study No.1: January 6, 2006.”

173



174 Part 6: Shockwave Swing Data

Figure 25-1 Swing chart, January 6, 2006.

Size 7,831

Open 1.2094

High 1.2182

Low 1.2087

Close 1.2162

Midrange 1.2135

Range 0.0095

Mean 1.2133


EURUSD 1/6/2006 Fri 0830 to 0859 ETSwing Chart Rev Amt = 15 Pips [EAIAIAHIAIAIAIEAI]

1.21821.21791.21761.21731.21701.21671.21641.21611.21581.21551.21521.21491.21461.21431.21401.21371.21341.21311.21281.21251.21221.21191.21161.21131.21101.21071.21041.21011.20981.20951.20921.2089

0830 0835 0840 0845 0850 0855


Swing Charts 175


Size 3,865

Open 1.2052

High 1.2090

Low 1.2039

Close 1.2063

Midrange 1.2065

Range 0.0051

Mean 1.2069


EURUSD 1/13/2006 Fri 0830 to 0859 ETSwing Chart Rev Amt = 10 Pips [IAIEAIE]

1.2089

1.2086

1.2083

1.2080

1.2077

1.2074

1.2071

1.2068

1.2065

1.2062

1.2059

1.2056

1.2053

1.2050

1.2047

1.2044

1.2041

0830 0835 0840 0845 0850 0855



Figure 25-3 Swing chart, January 20, 2006,

Size 2,352

Open 1.2069

High 1.2100

Low 1.2066

Close 1.2096

Midrange 1.2083

Range 0.0034

Mean 1.2084


EURUSD 1/20/2006 Fri 0830 to 0859 ETSwing Chart Rev Amt = 10 Pips [E]

1.2089

1.2095

1.2092

1.2089

1.2086

1.2083

1.2080

1.2077

1.2074

1.2071

1.2068

0830 0835 0840 0845 0850 0855


Swing Charts 177


Size 6,222

Open 1.2171

High 1.2238

Low 1.2170

Close 1.2220

Midrange 1.2204

Range 0.0068

Mean 1.2218


1.2238

1.2235

1.2232

1.2229

1.2226

1.2223

1.2220

1.2217

1.2214

1.2211

1.2208

1.2205

1.2202

1.2199

1.2196

1.2193

1.2190

1.2187

1.2184

1.2181

1.2178

1.2175

1.2172

0830 0835 0840 0845 0850 0855

EURUSD 1/27/2006 Fri 0830 to 0859 ET Swing Chart Rev Amt = 10 Pips [FIAIEAHIAHIEAHI]



Figure 25-5 Swing chart, February 3, 2006.

Size 7,772

Open 1.2062

High 1.2105

Low 1.1997

Close 1.2008

Midrange 1.2051

Range 0.0108

Mean 1.2032


1.21041.21011.20981.20951.20921.20891.20861.20831.20801.20771.20741.20711.20681.20651.20621.20591.20561.20531.20501.20471.20441.20411.20381.20351.20321.20291.20261.20231.20201.20171.20141.20111.20081.20051.20021.1999

0830 0835 0840 0845 0850 0855

EURUSD 2/3/2006 Fri 0830 to 0859 ETSwing Chart Rev Amt = 18 Pips [IEAIAHIAHI]


Swing Charts 179


Size 5,208

Open 1.1969

High 1.2023

Low 1.1962

Close 1.2018

Midrange 1.1993

Range 0.0061

Mean 1.1993


1.2021

1.2018

1.2015

1.2012

1.2009

1.2006

1.2003

1.2000

1.1997

1.1994

1.1991

1.1988

1.1985

1.1982

1.1979

1.1976

1.1973

1.1970

1.1967

1.1964

0830 0835 0840 0845 0850 0855

EURUSD 2/10/2006 Fri 0830 to 0859 ET Swing Chart Rev Amt = 10 Pips [IAIAIAIE]




Size 3,546

Open 1.1869

High 1.1881

Low 1.1858

Close 1.1875

Midrange 1.1870

Range 0.0023

Mean 1.1870


1.1881

1.1878

1.1875

1.1872

1.1869

1.1866

1.1863

1.1860

EURUSD 2/17/2006 Fri 0830 to 0859 ETSwing Chart Rev Amt = 10 Pips [IAHIEE]

0830 0835 0840 0845 0850 0855


Swing Charts 181


Size 3,833

Open 1.1898

High 1.1916

Low 1.1887

Close 1.1898

Midrange 1.1902

Range 0.0029

Mean 1.1901


1.1916

EURUSD 2/24/2006 Fri 0830 to 0859 ETSwing Chart Rev Amt = 10 Pips [HGDAIE]

1.1913

1.1910

1.1907

1.1904

1.1901

1.1898

1.1895

1.1892

1.1889

0830 0835 0840 0845 0850 0855



Figure 25-9 Swing chart, March 3, 2006.

Size 2,100

Open 1.2028

High 1.2041

Low 1.2023

Close 1.2038

Midrange 1.2032

Range 0.0018

Mean 1.2034


1.2040

1.2037

1.2034

1.2031

1.2028

1.2025

EURUSD 3/3/2006 Fri 0830 to 0859 ETSwing Chart Rev Amt = 10 Pips [I]

0830 0835 0840 0845 0850 0855


Swing Charts 183


Size 6,859

Open 1.1921

High 1.1925

Low 1.1885

Close 1.1909

Midrange 1.1905

Range 0.0040

Mean 1.1903


1.1923

1.1920

1.1917

1.1914

1.1911

1.1908

1.1905

1.1902

1.1899

1.1896

1.1893

1.1890

1.1887

0830 0835 0840 0845 0850 0855

EURUSD 3/10/2006 Fri 0830 to 0859 ETSwing Chart Rev Amt = 15 Pips [HHIAIAHIEAIAIEAI]




Size 3,393

Open 1.2187

High 1.2201

Low 1.2180

Close 1.2184

Midrange 1.2191

Range 0.0021

Mean 1.2190


EURUSD 3/17/2006 Fri 0830 to 0859 ETSwing Chart Rev Amt = 10 Pips [HIE]

1.2201

1.2199

1.2197

1.2195

1.2193

1.2191

1.2189

1.2187

1.2185

1.2183

1.2181

0830 0835 0840 0845 0850 0855


Swing Charts 185


Size 4,863

Open 1.1963

High 1.1990

Low 1.1951

Close 1.1968

Midrange 1.1971

Range 0.0039

Mean 1.1971


1.1990

1.1988

1.1986

1.1984

1.1982

1.1980

1.1978

1.1976

1.1974

1.1972

1.1970

1.1968

1.1966

1.1964

1.1962

1.1960

1.1958

1.1956

1.1954

1.1952

0830 0835 0840 0845 0850 0855

EURUSD 3/24/2006 Fri 0830 to 0859 ETSwing Chart Rev Amt = 10 Pips [ICDAIEAIE]




Size 1,681

Open 1.2113

High 1.2118

Low 1.2106

Close 1.2107

Midrange 1.2112

Range 0.0012

Mean 1.2111


0830 0835 0840 0845 0850 0855

1.2117

1.2115

1.2113

1.2111

1.2109

1.2107

EURUSD 3/31/2006 Fri 0830 to 0859 ETSwing Chart Rev Amt = 10 Pips [I]


Swing Charts 187

Figure 25-14 Swing chart, April 7, 2006.

Size 8,669

Open 1.2190

High 1.2227

Low 1.2171

Close 1.2194

Midrange 1.2199

Range 0.0056

Mean 1.2205


EURUSD 4/7/2006 Fri 0830 to 0859 ETSwing Chart Rev Amt = 15 Pips [IEAHIEE]

1.22261.22241.22221.22201.22181.22161.22141.22121.22101.22081.22061.22041.22021.22001.21981.21961.21941.21921.21901.21881.21861.21841.21821.21801.21781.21761.21741.2172

0830 0835 0840 0845 0850 0855



Figure 25-15 Swing chart, April 14, 2006.

Size 726

Open 1.2111

High 1.2115

Low 1.2103

Close 1.2110

Midrange 1.2109

Range 0.0012

Mean 1.2110


EURUSD 4/14/2006 Fri 0830 to 0859 ETSwing Chart Rev Amt = 10 Pips [HI]

1.2114

1.2112

1.2110

1.2108

1.2106

1.2104

0830 0835 0840 0845 0850 0855


See Table 25-1.

Swing Charts 189


Study Date Size High Low Range Std Dev Cycle ID

1 1/6/2006 7,831 1.2182 1.2087 95 25 EAIAIAHIAIAIAIEAI

2 1/13/2006 3,865 1.2090 1.2039 51 10 IAIEAIE

3 1/20/2006 2,352 1.2100 1.2066 34 8 E

4 1/27/2006 6,222 1.2238 1.2170 68 10 FIAIEAHIAHIEAHI

5 2/3/2006 7,772 1.2105 1.1997 108 17 IEAIAHIAHI

6 2/10/2006 5,208 1.2023 1.1962 61 16 IAIAIAIE

7 2/17/2006 3,546 1.1881 1.1858 23 5 IAHIEE

8 2/24/2006 3,833 1.1916 1.1887 29 6 HGDAIE

9 3/3/2006 2,100 1.2041 1.2023 18 3 I

10 3/10/2006 6,859 1.1925 1.1885 40 9 HHIAIAHIEAIAIEAI

11 3/17/2006 3,393 1.2201 1.2180 21 4 HIE

12 3/24/2006 4,863 1.1990 1.1951 39 8 ICDAIEAIE

13 3/31/2006 1,681 1.2118 1.2106 12 2 I

14 4/7/2006 8,669 1.2227 1.2171 56 10 IEAHIEE

15 4/14/2006 726 1.2115 1.2103 12 2 HI

Chapter 26Shockwave

Observations

OVERVIEW

The ideal shockwave is by definition the swing pattern that gen-erates the greatest profit. It would consist of a clearly defined surgewith only negligible retracements of less than 8 pips or so until itreaches its peak or valley. During this phase, traders hopefully willhave time to adjust trailing stop-loss orders online to protectearned gains, keeping the stop-loss price a distance of roughly 10to 15 pips away from the prevailing price.

Next will follow a very short period of small minor lateral move-ments indicating that the driving forces behind the price surge arebeginning to dissipate. This is equivalent to the change in the arith-metic sign of the slope of a trajectory.

Last but not least, there should be a reactive phase that movescontrary to the original outburst but on a more gradual slope.Traders then can exit voluntarily or allow their stop-loss orders tobe triggered. In the ideal shockwave, traders will reverse their posi-tions and ride the reaction phase back to the point of origin.

For the most part, the ideal shockwave with a complementaryretracement phase is a mythic concept. However, studies 2, 7, 10,11, and 14 do come close to approximating this ideal.

191


FRACTAL LEVELS

Swing data should be analyzed using different fractal levels. Thisis accomplished by incrementing or decrementing the minimumreversal amount in the swing reversal algorithm. In Figures 26-1and 26-2, I applied a 10- and a 15-pip reversal amount to the sameunderlying data.

An increase of just 5 pips in the reversal amount caused a dra-matic decrease in the number of diagonal waves, from 15 waves inFigure 26-1 to only 4 waves in Figure 26-2. The selection of mini-mum reversal amount is a function of the traders’ financial goals


1.2141

1.2138

1.2135

1.2132

1.2129

1.2126

1.2123

1.2120

1.2117

1.2114

1.2111

1.2108

1.2105

1.2102

1.2099

1.2096

0830 0831 0832

EURUSD 1/6/2006 Fri 0830 to 0831 ETSwing Chart (Rev Amt = 10 Pips)

Figure 26-1 Ten-pip minimum reversal amount.

and their propensity to the risk factor involved. The minimumreversal amount also assists traders in determining where to set thestop loss and take profit limit orders.

MAXIMA AND MINIMA

I will assume arbitrarily that if the causal forces are in fact present(i.e., several announcements were released on time, and they didnot cancel each other out), the resulting shockwave and its corres-ponding reaction will both occur between 8:30 and 9:00 a.m. Table26-1 lists the times when both extremes occurred for all 15 studies.

Shockwave Observations 193

1.2141

1.2138

1.2135

1.2132

1.2129

1.2126

1.2123

1.2120

1.2117

1.2114

1.2111

1.2108

1.2105

1.2102

1.2099

1.2096

0830 0831 0832

EURUSD 1/6/2006 Fri 0830 to 0831 ETSwing Chart (Rev Amt = 15 Pips)

Figure 26-2 Fifteen-pip minimum reversal amount.

Strangely enough, the standard deviation of the maximum valuesis 24.8 pips, whereas the standard deviation of the minimum valuesis only 16.0 pips. There is a distinct statistical tendency for shockwaveuptrends (the maxima) to diverge higher from the 8:30 a.m. open-ing price than there is to diverge lower in shockwave downtrends(the minima). However, this assumption may be slightly prematurebecause the sample size is only 15 data studies.

SWING VELOCITY

Swing velocity is the speed at which prices travel from one diagonalvertex to the next diagonal vertex (see Figure 26-3).

Each reversal point in a swing reversal chart (a peak or a valley)can be expressed in terms of Cartesian coordinate x, y pairs. InFigure 26-1, the swing diagonal begins at x1, y1 and ends at x2, y2.The slope of the diagonal line determines its swing velocity.Mathematically, this is described as the change in price divided bythe time elapsed or

Slope � (y2 � y1)/(x2 � x1)


TABLE 26-1 Maxima and Minima Amounts and Times

Study Date Maximum Max Time Minimum Min Time

1 1/6/2006 87 0853 �8 08372 1/13/2006 40 0841 �11 08313 1/20/2006 29 0858 �5 08344 1/27/2006 67 0843 �5 08305 2/3/2006 43 0831 �65 09006 2/10/2006 51 0857 �10 08317 2/17/2006 11 0835 �12 08238 2/24/2006 19 0836 �10 08479 3/3/2006 13 0858 �5 0831

10 3/10/2006 3 0830 �37 085011 3/17/2006 16 0850 �5 085612 3/24/2006 24 0841 �15 085213 3/31/2006 3 0831 �9 085114 4/7/2006 42 0833 �14 083115 4/14/2006 4 0848 �8 0859

Price is measured in pips, and time is measured in ticks. Onreexamination of the swing charts in Chapter 25, I find that theslope ranged from �48 (study no 6) to –23 (study no. 7). Theabsolute value of the slope of each diagonal is a valid indicator ofthe volatility during that time interval. High slope values indicateperiods of high profitability (or loss). Low slope values indicateperiods of lateral congestion.

Optionally, swing velocity can be “normalized” as the output ofa simple trigonometric function:

Θ � arctan(slope)

where values for angle Θ range from less than �90 to greater than–90 degrees.

BLIND TRADING

I coined this rather ominous sounding phrase to describe a tradingstrategy that may not be suitable to all traders’ dispositions. The blind


x2,y2

x2,y1x1,y1

Figure 26-3 Swing velocity.

modifier indicates that the trader is entering the market with a mini-mum of fundamental information on the current market situation.In the preceding 15 studies, I determined that the Friday, 8:30 a.m.shockwave rallied 85.7 percent of the time (see “Observations” inChapter 23).

The blind trading strategy can be tested using the demo accountthat nearly all currency dealers provide for their clients. The stepsare

• At 8:29 a.m., enter the market long.

• Set the stop-loss limit order at 10 to 12 pips below the marketentry price.

• Place a take-profit limit order at 45 pips above the market entryprice.

• If the prevailing price rises to 15 pips above the market entryprice, raise the stop-loss limit order to 5 pips above the marketentry price. If the market makes it this far, then a no-loss scenariohas been achieved.

• If the prevailing price advances to 25 pips above the marketentry price, raise the stop-loss limit order to 15 pips above themarket entry price.

• If the prevailing price advances to 35 pips above the marketentry price, raise the stop-loss limit order to 25 pips above themarket entry price.

• If the prevailing market price makes it this far, then it is timeto adjust the take-profit limit order, which should be raised to60 pips above the market entry price.

Depending on the strength and altitude of the shockwave,traders may opt to exit the market manually or wait until the pre-vailing price triggers a trailing stop-loss limit order.

Many traders probably are pondering why set a take-profit limitorder at all. The reason is that it is a defense mechanism againstthe highly improbable but possible loss of the connection to theonline trading platform. This heart-breaking electrical/electronicmalfunction has disastrous repercussions and has happened onlytwice since I began trading spot currencies in 2001. The first time


was a home system failure, and the second time was an incrediblybrief breakdown at my Internet service provider.

During a power outage anywhere along the communicationslink, it is possible that the market will trigger the take-profit limitorder and reverse back to the market entry point or lower. This iswhy traders should never initiate a new trade without both a stop-loss and a take-profit limit order. When setting a take-profit limitorder, make certain that it is a realistic distance from the prevail-ing price (usually less than 45 pips in the EURUSD currency pairduring high-volatility trading sessions).

RETRACEMENT TRADING

Using Newton’s law about an equal and opposite reaction for everyaction, many traders may be inclined to wait until the news shock-wave has declared its initial price direction. Then, at the shock-wave’s zenith or nadir, they enter the market in the oppositedirection. By examining the swing charts in Chapter 25, you cansee that it is possible to net 15 to 20 pips on the shockwave’s reac-tion phase. Again, this must be thoroughly tested via the dealer’sdemo account.

The trading prowess in retracement trading lies in the abilityto determine exactly when the shockwave has reached its peak orvalley. The maxima and minima table presented earlier may assisttraders in this assessment.

INFORMED TRADING

When traders have informed insight into the likelihood of thenews announcement numbers, the dilemma of which position to initiate is greatly reduced. Several news agencies and cable sta-tions broadcast timely opinions on predicted changes in pricedirectionality. Always recall that announcements that indicate abeneficial influence on the U.S. economy mean a decline in theEURUSD pair, and vice versa.


The strategy for “informed” trading is to monitor one or morenews agencies, financial narrators, or broadcast networks startingat 8:00 a.m. Maintain detailed records of the preshockwave fore-casts for each agency. At 10:00 a.m., display the EURUSD chart inthe currency dealer’s trading platform, and record the perfor-mance of each agency. After several weeks, traders will have a goodidea of whose forecasts warrant credibility and whose do not.

CAVEAT

I admit that the current work on news release shockwaves is not anoverly exhaustive compendium of detailed research on the topic.I confined my studies to the single currency pair EURUSD and thetime frame 8:00 to 10:00 a.m. on Fridays. Nonetheless, it does pro-vide traders with a very diversified introduction to trading the mostvolatile time frames in the world’s largest financial market.


Appendices


Appendix AISO Currency Pairs

This is a list of global currencies and the three-character currencycodes that I have found are generally used to represent them.Often, but not always, this code is the same as the ISO 4217 stan-dard. (The ISO, or International Organization for Standardization,is a worldwide federation of national standards.)

In most cases, the currency code is composed of the country’stwo-character Internet country code plus an extra character to denote the currency unit. For example, the code for Canadiandollars is simply Canada’s two-character Internet country code(CA) plus a one-character currency designator (D).

I have endeavored to list the codes that, in my experience, areactually in general industry use to represent the currencies.Currency names are given in the plural form. This list does notcontain obsolete Euro-zone currencies.

201

TABLE A-1 World Currencies

Symbol Region/Country Currency Name

AED United Arab Emirates DirhamsAFA Afghanistan AfghanisALL Albania LekeAMD Armenia DramsANG Netherlands Antilles GuildersAOA Angola KwanzaARS Argentina PesosAUD Australia Dollars

Cont.


202 Appendices


AWG Aruba GuildersAZM Azerbaijan ManatsBAM Bosnia, Herzegovina Convertible markaBBD Barbados DollarsBDT Bangladesh TakaBGN Bulgaria LevaBHD Bahrain DinarsBIF Burundi FrancsBMD Bermuda DollarsBND Brunei Darussalam DollarsBOB Bolivia BolivianosBRL Brazil Brazil realBSD Bahamas DollarsBTN Bhutan NgultrumBWP Botswana PulasBYR Belarus RublesBZD Belize DollarsCAD Canada DollarsCDF Congo/Kinshasa Congolese francsCHF Switzerland FrancsCLP Chile PesosCNY China RenminbiCOP Colombia PesosCRC Costa Rica ColonesCUP Cuba PesosCVE Cape Verde EscudosCYP Cyprus PoundsCZK Czech Republic KorunyDJF Djibouti FrancsDKK Denmark KronerDOP Dominican Republic PesosDZD Algeria Algeria dinarsEEK Estonia KrooniEGP Egypt PoundsERN Eritrea NakfaETB Ethiopia BirrEUR Euro member countries EuroFJD Fiji DollarsFKP Falkland Islands PoundsGBP United Kingdom PoundsGEL Georgia LariGGP Guernsey PoundsGHC Ghana Cedis

ISO Currency Pairs 203


GIP Gibraltar PoundsGMD Gambia DalasiGNF Guinea FrancsGTQ Guatemala QuetzalesGYD Guyana DollarsHKD Hong Kong DollarsHNL Honduras LempirasHRK Croatia KunaHTG Haiti GourdesHUF Hungary ForintIDR Indonesia RupiahsILS Israel New shekelsIMP Isle of Man PoundsINR India RupeesIQD Iraq DinarsIRR Iran RialsISK Iceland KronurJEP Jersey PoundsJMD Jamaica DollarsJOD Jordan DinarsJPY Japan YenKES Kenya ShillingsKGS Kyrgyzstan SomsKHR Cambodia RielsKMF Comoros FrancsKPW Korea (North) WonKRW Korea (South) WonKWD Kuwait DinarsKYD Cayman Islands DollarsKZT Kazakstan TengeLAK Laos KipsLBP Lebanon PoundsLKR Sri Lanka RupeesLRD Liberia DollarsLSL Lesotho MalotiLTL Lithuania LitaiLVL Latvia LatiLYD Libya DinarsMAD Morocco DirhamsMDL Moldova LeiMGA Madagascar AriaryMKD Macedonia DenarsMMK Myanmar (Burma) Kyats

Cont.

204 Appendices


MNT Mongolia TugriksMOP Macau PatacasMRO Mauritania OuguiyasMTL Malta LiriMUR Mauritius RupeesMVR Maldives RufiyaaMWK Malawi KwachasMXN Mexico PesosMYR Malaysia RinggitsMZM Mozambique MeticaisNAD Namibia DollarsNGN Nigeria NairasNIO Nicaragua Gold cordobasNOK Norway KroneNPR Nepal Nepal rupeesNZD New Zealand DollarsOMR Oman RialsPAB Panama BalboaPEN Peru Nuevos solesPGK Papua New Guinea KinaPHP Philippines PesosPKR Pakistan RupeesPLN Poland ZlotychPYG Paraguay GuaraniQAR Qatar RialsROL Romania LeiRUR Russia RublesRWF Rwanda Rwanda francsSAR Saudi Arabia RiyalsSBD Solomon Islands DollarsSCR Seychelles RupeesSDD Sudan DinarsSEK Sweden KronorSGD Singapore DollarsSHP Saint Helena PoundsSIT Slovenia TolarsSKK Slovakia KorunySLL Sierra Leone LeonesSOS Somalia ShillingsSPL Seborga LuiginiSRG Suriname GuildersSTD São Tome, Principe DobrasSVC El Salvador Colones

ISO Currency Pairs 205


SYP Syria PoundsSZL Swaziland EmalangeniTHB Thailand BahtTJS Tajikistan SomoniTMM Turkmenistan ManatsTND Tunisia DinarsTOP Tonga Pa’angaTRL Turkey LirasTTD Trinidad, Tobago DollarsTVD Tuvalu Tuvalu dollarsTWD Taiwan New dollarsTZS Tanzania ShillingsUAH Ukraine HryvniaUGX Uganda ShillingsUSD United States DollarsUYU Uruguay PesosUZS Uzbekistan SumsVEB Venezuela BolivaresVND Viet Nam DongVUV Vanuatu VatuWST Samoa TalaXAG Silver OuncesXAU Gold OuncesXCD East Caribbean DollarsXDR International Monetary Fund Special drawing rightsXPD Palladium OuncesXPT Platinum OuncesYER Yemen RialsYUM Yugoslavia New dinarsZAR South Africa RandZMK Zambia KwachaZWD Zimbabwe Zimbabwe dollars

Appendix BExchange Rates

Table B-1 shows the international foreign exchange rates on April21, 2006 compared with the USD.

It is interesting to note that as of this date, only six world curr-encies have a parity rate with the USD greater than 1.0000: Kuwaitidinar (3.42407), Baharaini dinar (2.65266), Omani rial (2.59774),Cypriot pound (2.14777), British pound (1.78280), and the euro(1.23450). Coincidentally, at the bottom of the list, both alphabet-ically and parity-wise, is the Zimbabwe dollar, which requires over99,000 to equal 1 USD.

Additional information on current exchange rates can be foundat http://moneycentral.msn.com/investor/market/rates.asp.

207

Cont.

TABLE B-1 International Foreign Exchange Rates, April 21, 2006

Currency Units/USD USD/Units

Algerian dinar 0.01379 72.52500Argentine peso 0.32701 3.05800Australian dollar 0.74420 1.34373Baharaini dinar 2.65266 0.37698Bolivian boliviano 0.12508 7.99500Brazilian real 0.47279 2.11510British pound 1.78280 0.56092Botswana pula 0.18714 5.34360Canadian dollar 0.87827 1.13860Chilean peso 0.00193 517.54999Chinese yuan 0.12477 8.01450


http://moneycentral.msn.com/investor/market/rates.asp

208 Appendices

Currency Units/USD USD/Units

Columbian peso 0.00043 2,337.00004Cypriot pound 2.14777 0.46560Czech koruna 0.04359 22.94200Danish krone 0.16547 6.04350Ecuador sucre 0.00004 25,000.00063Euro 1.23450 0.81005Ghanaian cedi 0.00011 9,106.99988Guatemalan quetzal 0.13201 7.57500Hong Kong dollar 0.12897 7.75400Hungarian forint 0.00467 213.96001Israeli shekel 0.22015 4.54230Indian rupee 0.02216 45.13500Indonesian rupiah 0.00011 8,882.99974Japanese yen 0.00855 116.93001Jordanian dinar 1.41143 0.70850Kenyan shilling 0.01404 71.22000Kuwaiti dinar 3.42407 0.29205Malaysian ringgit 0.27319 3.66050Mexican peso 0.09016 11.09120Moroccan dirham 0.11191 8.93550Namibian dollar 0.16587 6.02900New Zealand dollar 0.63330 1.57903Norwegian krone 0.15748 6.35000Omani rial 2.59774 0.38495Pakistan rupee 0.01668 59.97000Peruvian nuevo sol 0.30233 3.30770Qatari rial 0.27467 3.64070Russian ruble 0.03639 27.48000Saudi riyal 0.26663 3.75050Singapore dollar 0.62661 1.59590South African rand 0.16707 5.98550South Korean won 0.00106 948.00005Swedish krona 0.13248 7.54860Swiss franc 0.78475 1.27430Taiwan dollar 0.03098 32.27500Tanzanian shilling 0.00083 1,211.99996Thai baht 0.02645 37.81000Tunisian dinar 0.74738 1.33800Turkish lira 0.75683 1.32130UAR emirati dirham 0.27228 3.67270U.S. dollar 1.00000 1.00000Venezualan bolivar 0.00047 2,144.00005Vietnamese dong 0.00006 15,924.99916Zimbabwe dollar 0.00001 99,202.00100

Appendix CEuro Currency

On January 1, 1999, eleven of the countries in the EuropeanEconomic and Monetary Union (EMU) decided to give up theirown currencies and adopt the new euro (EUR) currency: Austria,Belgium, Finland, France, Germany, Ireland, Italy, Luxembourg,the Netherlands, Portugal, and Spain. Greece followed suit onJanuary 1, 2001. Vatican City also participated in the changeover.This changeover is now complete.

It is worth noting that any place that previously used one or moreof the currencies listed below has now also adopted the euro. Thisapplies to the Principality of Andorra, the Principality of Monaco,and the Republic of San Marino. This, of course, applies automat-ically to any territories, departments, possessions, or collectivitiesof Euro-zone countries, such as the Azores, Balearic Islands, theCanary Islands, Europa Island, French Guiana, Guadeloupe, Juande Nova, the Madeira Islands, Martinique, Mayotte, Reunion, Saint-Martin, Saint Pierre, and Miquelon, to name just a few.

Euro bank notes and coins began circulating in these countrieson January 1, 2002. At that time, all transactions in those countrieswere valued in euros, and the “old” notes and coins of these coun-tries were gradually withdrawn from circulation. The precise datesthat each “old” currency ceased being legal tender are noted inTable C-1.

For convenience, and because their values are now irrevocablyset against the euro as listed above, the XE.com Universal CurrencyConverter will continue to support these units even after their

209


withdrawal from circulation. In addition, most outgoing Euro currencies still will be physically convertible at special locations fora period of several years. For details, refer to the official Euro site.

Also note that the euro is not just the same thing as the formerEuropean Currency Unit (or ECU), which used to be listed asXEU. The ECU was a theoretical “basket” of currencies rather thana currency in and of itself, and no ECU bank notes or coins everexisted. At any rate, the ECU has been replaced by the euro, whichis a bona fide currency.

A note about spelling and capitalization: The official spelling ofthe EUR currency unit in the English language is euro, with a low-ercase e. However, the overwhelmingly prevailing industry practiceto spell it Euro, with a capital E. Since other currency names are cap-italized in general use, doing so helps to differentiate the nounEuro, meaning EUR currency, from the more general adjective euro,meaning anything even remotely having to do with Europe.

210 Appendices

TABLE C-1 Official Fixed Euro Rates for Participating Countries

Symbol Country Legacy (Old) Conversion to Conversion Currency Euro from Euro

ATS Austria Schilling ATS/13.7603 � EUR EUR � 13.7603 � ATSBEF Belgium Franc BEF/40.3399 � EUR EUR � 40.3399 � BEFDEM Germany Mark DEM/1.95583 � EUR EUR � 1.95583 � DEMESP Spain Peseta ESP/166.386 � EUR EUR � 166.386 � ESPFIM Finland Markka FIM/5.94573 � EUR EUR � 5.94573 � FIMFRF France Franc FRF/6.55957 � EUR EUR � 6.55957 � FRFGRD Greece Drachma GRD/340.750 � EUR EUR � 340.750 � GRDIEP Ireland Punt IEP/0.787564 � EUR EUR � 0.787564 � IEPITL Italy Lira ITL/1936.27 � EUR EUR � 1936.27 � ITLLUF Luxembourg Franc LUF/40.3399 � EUR EUR � 40.3399 � LUFNLG Netherlands Guilder NLG/2.20371 � EUR EUR � 2.20371 � NLGPTE Portugal Escudo PTE/200.482 � EUR EUR � 200.482 � PTEVAL Vatican City Lira VAL/1936.27 � EUR EUR � 1936.27 � VAL

Appendix DGlobal Banking

Hours

Price fluctuations in the spot currency markets are essentiallynews-driven. Or more accurately, it is the human reaction to news-driven events that makes trading possible and profitable. Howtraders interpret these news events determines which direction themarket will travel. As in all financial markets, the foreign exchangealso has its share of contrarians who keep runaway breakouts incheck while supplying additional volatility to the overall situation.

Despite all the fundamental and technical influences on the for-eign exchange, one major constant in determining periods of highvolatility is the hours of operation for the central banks of eachmajor currency country.

Figure D-1 emphasizes the importance of the effect of time ofday on Forex market activity and volatility based on hours of oper-ation around the globe. Because banking hours vary from coun-try to country, I have arbitrarily set hours of operation from 9:00a.m. to 5:00 p.m. for consistency. The top row is expressed asCentral European Time (Greenwich Mean Time � 1 hour), whichaligns with the Central Bank of Europe in Frankfurt, the most pres-tigious central bank in the European Monetary Union.

This figure allows traders to view overlapping time periodswhen central banks for different currencies are operating and thusguarantee a certain degree of mutual activity.

211


For example, when banks open in New York City at 9:00 a.m.EST, the Frankfurt bank has already been operating for six hours.Thus there is a two-hour overlap of trading in the EURUSD currency pair on both sides of the Atlantic Ocean (9:00 to 11:00a.m. EST). This can be readily recognized in the time-of-day activ-ity chart for the EURUSD pair.

If you are interested in initiating a trade in the EURHKD cross-rate pair, note that there is a one-hour overlap in banking oper-ations between central Europe and Hong Kong that occursbetween 9:00 and 10:00 a.m. in Frankfurt (or 3:00 and 4:00 a.m.in New York).

Dedicated currency traders may have to adjust their sleepingschedules to take advantage of increased activity and volatility whentrading non-USD cross-rate currency pairs.

BANK HOLIDAYS

In Table D-1, bank holidays are sorted by date and referenced bythe ISO currency name. The euro does not appear in the list, butthe three most industrialized component members are representedas Germany (DEM), France (FRF), and Italy (ITL).

212 Appendices

EUR

CHF

CBP

BKL

USD

CAD

NZD

AUD

JPY

HKD

SCD

THB

RUR

ZAK

Euro Currency

Swiss Franc

British Pound

Brarilian Real

United States Dollar

Canadian Dollar

New Zealand Dollar

Australian Dollar

Japanese Yen

Hong Kong Dollar

Singapore Dollar

Thai Bahr

Russian Ruble

South African Rasd

Frankfurt

Zurich

London

Kis de Janeire

New York

Montreal

Wellington

Sydney

Tokyo

Hong Kong

Singapore

Bangkok

Mascow

Jehannesburg

GLOBAL BANKING HOURSBased on Central European Time (GMT+1)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

Figure D-1 Global banking hours.

Global Banking Hours 213

TABLE D-1 International Bank Holidays

Date Holiday Currencies

1/1/2006 New Year’s Day AUD, CAD,CHF,DEM,FRF, GBP, HKD, ITL,JPY, SGD,USD

1/2/2006 New Year’s Holiday JPY

1/3/2006 New Year’s Holiday AUD, JPY

1/6/2006 Epiphany ITL

1/9/2006 Coming of Age Day JPY

1/16/2006 Martin Luther King’s Day USD

1/21/2006 Hari Raya Haji SGD

1/26/2006 Australia Day AUD

1/28/2006 Chinese New Year HKD

1/30/2006 Chinese New Year HKD, SGD

1/31/2006 Chinese New Year HKD, SGD

2/11/2006 National Foundation Day JPY

2/21/2006 President’s Day USD

3/6/2006 Labor Day AUD

3/21/2006 Spring Equinox JPY

4/4/2006 Tomb-Sweeping Day HKD

4/5/2006 Tomb-Sweeping Day HKD

4/14/2006 Good Friday AUD, CAD, CHF,DEM, GBP, HKD, SGD

4/17/2006 Easter Monday AUD, CHF,DEM, FRF, GBP, HKD, ITL

4/25/2006 ANZAC Day AUD

4/29/2006 Day of Nature JPY

5/1/2006 Labor Day DEM, FRF, HKD, ITL, SGD

5/1/2006 Early May Bank Holiday GBP

5/3/2006 Constitution Day JPY

5/4/2006 Citizen’s Day of Rest JPY

5/5/2006 Buddha’s Anniversary HKD

5/5/2006 Children’s Day JPY

5/8/2006 Liberation Day FRF

5/22/2006 Buddha Purnima SGD

5/22/2006 Victoria Day CAD

5/25/2006 Ascension Thursday CHF,DEM, FRF

5/29/2006 Memorial Day USD

5/29/2006 Spring Bank Holiday GBP

5/31/2006 Dragon Boats Festival HKD

6/2/2006 Republic’s Day ITL

6/4/2006 Pentecost Monday FRF

Cont.

214 Appendices

Date Holiday Currencies

6/5/2006 Whit Monday CHF,DEM

6/12/2006 Queen’s Birthday AUD

6/22/2006 Lord’s Day CHF

6/29/2006 St. Peter & St. Paul CHF

7/1/2006 Special Administration Day HKD

7/1/2006 Canada Day CAD

7/4/2006 Independence Day USD

7/14/2006 Bastille Day FRF

7/18/2006 Sea’s Day JPY

7/20/2006 Navy Day JPY

8/1/2006 Confederation Day CHF

8/9/2006 National Day SGD

8/15/2006 Assumption FRF,ITL

8/28/2006 Summer Bank Holiday GBP

9/4/2006 Labor Day CAD

9/15/2006 Respect for the Aged Day JPY

9/23/2006 Autumn Equinox JPY

10/1/2006 National Day HKD

10/3/2006 National Day DEM

10/7/2006 Late Mid-Autumn Festival HKD

10/9/2006 Physical Fitness Day JPY

10/9/2006 Columbus Day USD

10/30/2006 Chung Yeung Festival HKD

11/1/2006 All Saint’s Day FRF,ITL

11/1/2006 Dipabali SGD

11/3/2006 National Culture Day JPY

11/3/2006 Hari Raya Puasa SGD

11/11/2006 Armistice Day FRF

11/11/2006 Veteran’s Day USD

11/11/2006 Remembrance Day CAD

11/23/2006 Labor Thanksgiving Day JPY

11/23/2006 Thanksgiving Day USD

12/8/2006 Immaculate Conception CHF,ITL

12/23/2006 Emperor’s Birthday JPY

12/25/2006 Christmas Day AUD,CAD,CHF,FRF,GBP,HKD,ITL,SGD,USD

12/26/2006 St. Stephen’s Day ITL

12/26/2006 Boxing Day AUD,CAD,DEM,GBP,HKD

12/31/2006 Bank Holiday JPY

12/31/2006 New Year’s Eve DEM,USD

Boxing Day is a public holiday observed in many Commonwealthcountries on the first day other than Sunday following Christmas Day.

Whit Monday is a Christian holiday celebrated the day afterPentecost, a movable feast in the Christian calendar depending onthe date of Easter.

ANZAC Day is commemorated in Australia and New Zealand tohonor the bravery and sacrifice of the Australian and New ZealandArmy Corps (ANZAC).

Most CFTC-registered currency dealers provide their clientswith some of the preceding information online. Investors who arepartial to trading the Pacific Rim currency pairs always should con-sult their dealer’s holiday calendars. The reason is that Japan cel-ebrates 18 bank holidays every year, and Singapore has 17 annualbank holidays (the United Kingdom celebrates only 8 bank holi-days a year). Banking holidays characteristically signify a decreasein trading volume and liquidity. In addition, slippage (when ordersare filled at prices other than those designated by the trader) isprone to occur during holidays and periods of low liquidity.

Global Banking Hours 215

Appendix EMonthly OHLC

and ActivityCharts

In this appendix I present monthly OHLC and activity charts forthe 10 most frequently traded USD currency pairs. Beneath eachchart are the corresponding statistics for the time period January2000 through December 2005.

In the statistics block, I will use the following definitions:

Midrange � (high � low)/2Absolute range � high � low

Relative range � 100 � absolute range/midrange

The currency pairs in this appendix are arranged in the fol-lowing order :

EURUSD

GBPUSD

USDCHF

USDJPY

USDCAD

AUDUSD

217


NZDUSD

USDSEK

USDNOK

USDDKK

In the charts that follow, activity is represented as the vertical barsat the bottom. The upper (lighter) portion is the number of upticks,whereas the lower (darker) portion is the number of downticks.Their sum equals the total activity for the corresponding month.

218 Appendices

Monthly OHLC and Activity Charts 219

1.3300

1.2800

1.2300

1.1800

1.1300

1.0800

1.0300

0.9800

0.9300

0.8800

0.8300

2,750,0002,500,0002,250,0002,000,0001,750,0001,500,0001,250,0001,000,000

750,000500,000250,000

0

EURUSD - Jan 2000 to Dec 2005 Monthly OHLC and Activity

J FMAMJ J ASONDJ FMAMJ J ASONDJ FMAM JJ ASONDJ FMA M J J A SONDJ FMAM J J A SONDJ FMAMJ J ASOND

Figure E-1 EURUSD monthly OHLC and activity.

Open 1.0088High 1.3667Low 0.8229Close 1.1849Mean 1.0656Midrange 1.0948Absolute range 0.5438Relative range 49.6712Standard deviation 0.1573

It is interesting to note that activity increased by 1,060 percentfrom January 2000 to December 2005.

220 Appendices

1.9300

1.8800

1.8300

1.7800

1.7300

1.6800

1.6300

1.5800

1.5300

1.4800

1.4300

1.3800

2,750,0002,500,0002,250,0002,000,0001,750,0001,500,0001,250,0001,000,000

750,000500,000250,000

0J FMAMJ J A SOND J FMAMJ J A SOND J MAMJ J A SOND J FMAMJ J A SOND J FMAMJ J A SOND J FMAMJ J A SONDF

GBPUSD - Jan 2000 to Dec 2005Monthly OHLC and Activity

Figure E-2 GBPUSD monthly OHLC and activity.



D

1.79001.74001.6900

1.64001.5900

1.54001.49001.4400

1.39001.34001.2900

1.2400

1.19001.1400

2,750,0002,500,0002,250,0002,000,0001,750,0001,500,0001,250,0001,000,000

750,000500,000250,000

0J FMAMJ J A SOND J FMAMJ J A SOND J MAMJ J A SOND J FMAMJ J A SOND J FMAMJ J A SOND J FMAMJ J A SONF

USDCHF - Jan 2000 to Dec 2005Monthly OHLC and Activity

Figure E-3 USDCHF monthly OHLC and activity.


222 Appendices

134.50

132.00

129.50

127.00

124.50

122.00

119.50

117.00

114.50

112.00

109.50

107.00

104.50

102.00

2,250,000

2,000,000

1,750,000

1,500,000

1,250,000

1,000,000

750,000

500,000

250,000

0

USDJPY - Jan 2000 to Dec 2005Monthly OHLC and Activity

DJ FMAMJ J A SOND J FMAMJ J A SOND J MAMJ J A SOND J FMAMJ J A SOND J FMAMJ J A SOND J FMAMJ J A SONF

Figure E-4 USDJPY monthly OHLC and activity.




USDCAD - Jan 2000 to Dec 2005Monthly OHLC and Activity

1.60001.57501.55001.52501.50001.47501.45001.42501.40001.37501.35001.32501.30001.27501.25001.22501.20001.17501.1500

1,375,0001,250,0001,125,0001,000,000

875,000750,000625,000500,000375,000250,000125,000

0

Figure E-5 USDCAD monthly OHLC and activity.


224 Appendices


0.78500.76000.73500.71000.68500.66000.63500.61000.58500.56000.53500.51000.4850

1,375,0001,250,0001,125,0001,000,000

875,000750,000625,000500,000375,000250,000125,000

0

AUDUSD - Jan 2000 to Dec 2005Monthly OHLC and Activity

Figure E-6 AUDUSD monthly OHLC and activity.




0.72500.70000.67500.65000.62500.60000.57500.55000.52500.50000.47500.45000.42500.4000

1,125,0001,000,000

875,000750,000625,000500,000375,000250,000125,000

0

NZDUSD - Jan 2000 to Dec 2005Monthly OHLC and Activity

Figure E-7 NZDUSD monthly OHLC and activity.


226 Appendices

10.85010.60010.35010.100

9.8509.6009.3509.1008.8508.6008.3508.1007.8507.6007.3507.1006.8506.600

1,125,000

1,000,000

875,000

750,000

625,000

500,000

375,000

250,000

125,000

0DJ FMAMJ J A SOND J FMAMJ J A SOND J MAMJ J A SOND J FMAMJ J A SOND J FMAMJ J A SOND J FMAMJ J A SONF

USDSEK - Jan 2000 to Dec 2005Monthly OHLC and Activity

Figure E-8 USDSEK monthly OHLC and activity.



9.300

8.5508.3008.0507.8007.5507.3007.0506.8006.5506.3006.050

1,000,000

875,000

750,000

625,000

500,000

375,000

250,000

125,000

0

8.8009.050

9.550


USDNOK - Jan 2000 to Dec 2005Monthly OHLC and Activity

Figure E-9 USDNOK monthly OHLC and activity


228 Appendices


8.9508.7008.4508.2007.9507.7007.4507.2006.9506.7006.4506.2005.9505.7005.450

875,000

750,000

625,000

500,000

375,000

250,000

125,000

0

USDDKK - Jan 2000 to Dec 2005Monthly OHLC and Activity

Figure E-10 USDDKK monthly OHLC and activity

Open 7.3780High 9.0425Low 5.4443Close 6.2978Mean 7.1353Midrange 7.2434Absolute range 3.5982Relative Range 49.6756Standard deviation 1.0510

Appendix FDaily OHLC andActivity Charts

This appendix is analogous to the preceding one, except that thetime frame spans January 1, 2006 through April 14, 2006, and thetime interval has been changed from monthly to daily in order toscrutinize recent market characteristics in greater detail.

The numeric values in the statistical blocks beneath each chartare defined as in Appendix E.

The currency pairs in this appendix are arranged in the fol-lowing order :

EURUSD

GBPUSD

USDCHF

USDJPY

USDCAD

AUDUSD

NZDUSD

USDSEK

USDNOK

USDDKK

229


230 Appendices

1.2325

EURUSD 1/1/2006 to 4/14/2006Daily OHLC and Activity

1.23001.22751.22501.22251.22001.21751.21501.21251.21001.20751.20501.20251.20001.1975

1.19251.1950

1.19001.18751.18501.1825

180,000160,000140,000120,000100,00080,00060,00040,00020,000

SMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTF0

Figure F-1 EURUSD daily OHLC and activity.


Daily OHLC and Activity Charts 231

Figure F-2 GBPUSD daily OHLC and activity.


1.7900

GBPUSD 1/1/2006 to 4/14/2006Daily OHLC and Activity

1.7850

1.7800

1.7750

1.7700

1.7650

1.7600

1.7550

1.7500

1.7450

1.7400

1.7350

1.7300

1.7250

1.7200200,000175,000150,000125,000100,00075,00050,00025,000

0SMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTF

232 Appendices

1.3200

1.3150

1.3100

1.3050

1.3000

1.2950

1.2900

1.2850

1.2800

1.2750

1.2700

1.2650

1.2600

USDCHF 1/1/2006 to 4/14/2006Daily OHLC and Activity

SMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTF

175,000150,000125,000100,00075,00050,00025,000

0

Figure F-3 USDCHF daily OHLC and activity.



Figure F-4 USDJPY daily OHLC and activity.


119.00

118.50

118.00

117.50

117.00

116.50

116.00

115.50

115.00

114.50

114.00

USDJPY 1/1/2006 to 4/14/2006Daily OHLC and Activity


175,000150,000125,000100,00075,00050,00025,000

0

234 Appendices

1.18001.17751.17501.17251.17001.16751.16501.16251.16001.15751.15501.15251.15001.14751.14501.14251.14001.13751.13501.1325

USDCAD 1/1/2006 to 4/14/2006Daily OHLC and Activity


100,00080,00060,00040,00020,000

0

Figure F-5 USDCAD daily OHLC and activity.



AUDUSD 1/1/2006 to 4/14/2006Daily OHLC and Activity

.7575

.7550

.7525

.7500

.7475

.7450

.7425

.7400

.7375

.7350

.7325

.7300

.7275

.7250

.7225

.7200

.7175

.7150

.7125

.7100

.7075

.7050

100,00080,00060,00040,00020,000

0SMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTF

Figure F-6 AUDUSD daily OHLC and activity.


236 Appendices

NZDUSD 1/1/2006 to 4/14/2006Daily OHLC and Activity


.6990

.6960

.6930

.6900

.6870

.6840

.6810

.6780

.6750

.6720

.6690

.6660

.6630

.6600

.6570

.6540

.6510

.6480

.6450

.6420

.6390

.6360

.6330

.6300

.6270

.6240

.6210

.6180

.6150

.6120

.6090

.6060

.6030

100,000

80,000

60,000

40,000

20,000

0

Figure F-7 NZDUSD daily OHLC and activity.



7.97007.94507.92007.89507.87007.84507.82007.79507.77007.74507.72007.69507.67007.64507.62007.59507.57007.54507.5200

USDSEK 1/1/2006 to 4/14/2006Daily OHLC and Activity


100,000

80,000

60,000

40,000

20,000

0

Figure F-8 USDSEK daily OHLC and activity.


238 Appendices

6.8550

6.83006.80506.78006.75506.73006.70506.68006.65506.63006.60506.58006.55506.53006.50506.48006.45506.43006.4050


USDNOK 1/1/2006 to 4/14/2006Daily OHLC and Activity

120,000100,00080,00060,00040,00020,000

0

Figure F-9 USDNOK daily OHLC and activity.



6.32006.31006.30006.29006.28006.27006.26006.25006.24006.23006.22006.21006.20006.19006.18006.17006.16006.15006.14006.13006.12006.11006.10006.09006.08006.07006.0600

75,000

60,000

45,000

30,000

15,000


USDDKK 1/1/2006 to 4/14/2006Daily OHLC and Activity

0

Figure F-10 USDDKK daily OHLC and activity.


Appendix GComposite

Activity Charts(Daily)

The time frame in the following charts spans from January 1, 2005through April 14, 2006. Composite charts are calculated by aver-aging the sum of the upticks and downticks over that period using1-minute time intervals. Their purpose is to assist traders in deter-mining when to schedule online trading sessions based on traders’predilection for the nebulous risk/reward factor and the volatilityof the targeted currency pair (in this case, EURUSD).

The vertical numeric scale on the right of each chart is activityexpressed in total number of ticks (upticks plus downticks) dur-ing each time interval. The bottom band (the darkest) representsthe activity for the current minute. The central band plus the lowerband represents 3-minute activity. The sum of the all three verti-cal bars represents the 5-minute activity.

241


242 Appendices

EURUSD Time of Day Chart (Sundays)1-, 3- and 5-Minute Activity (1/1/2005 to 4/14/2006)

2400230022002100200019001800170016001500140013001200110010000900080007000600050004000300020001000000

200

190

180

170

160

150

140

130

120

110

100

90

80

70

60

50

40

30

20

10

0

Figure G-1 Sunday composite activity chart.

Composite Activity Charts (Daily) 243

EURUSD Time of Day Chart (Mondays)1-, 3- and 5-Minute Activity (1/1/2005 to 4/14/2006)

2400230022002100200019001800170016001500140013001200110010000900080007000600050004000300020001000000

4204104003903803703603503403303203103002902802702602502402302202102001901801701601501401301201101009080706050403020100

Figure G-2 Monday composite activity chart.

244 Appendices

EURUSD Time of Day Chart (Tuesdays)1-, 3- and 5-Minute Activity (1/1/2005 to 4/14/2006)

2400230022002100200019001800170016001500140013001200110010000900080007000600050004000300020001000000

5105004904804704604504404304204104003903803703603503403303203103002902802702602502402302202102001901801701601501401301201101009080706050403020100

Figure G-3 Tuesday composite activity chart.


EURUSD Time of Day Chart (Wednesdays)1-, 3- and 5-Minute Activity (1/1/2005 to 4/14/2006)

2400230022002100200019001800170016001500140013001200110010000900080007000600050004000300020001000000

500490480470460450440430420410400390380370360350340330320310300290280270260250240230220210200190180170160150140130120110100

9080706050403020100

Figure G-4 Wednesday composite activity chart.

246 Appendices

EURUSD Time of Day Chart (Thursdays)1-, 3- and 5-Minute Activity (1/1/2005 to 4/14/2006)

2400230022002100200019001800170016001500140013001200110010000900080007000600050004000300020001000000

6156005855705555405255104954804654504354204053903753603453303153002852702552402252101951801651501351201059075604530150

Figure G-5 Thursday composite activity chart.


EURUSD Time of Day Chart (Fridays)1-, 3- and 5-Minute Activity (1/1/2005 to 4/14/2006)

2400230022002100200019001800170016001500140013001200110010000900080007000600050004000300020001000000

5855705555405255104954804654504354204053903753603453303153002852702552402252101951801651501351201059075604530150

Figure G-6 Friday composite activity chart.

Appendix HComposite Range

Charts (Daily)

The time frame in the following charts spans from January 1, 2005through April 14, 2006.

The vertical numeric scale on the right of each chart is absoluterange expressed in integer pips of the quote currency. The bot-tom band (the darkest) represents the range for the current 2-minute interval. The sum of the central band plus the lower bandrepresents 7-minute range. The sum of all three vertical bands rep-resents the 15-minute absolute range.

249


250 Appendices

0000 0100 0200 0300 0400 0500 0600 0700 0800 0900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000 2100 2200 2300 2400

15

14

13

12

11

10

9

8

7

6

5

4

3

2

1

0

EURUSD Time of Day Chart (Sundays)2-, 7- and 15-Minute Range (1/1/2005 to 4/14/2006)

Figure H-1 Sunday composite range chart.

Composite Range Charts (Daily) 251

0000 0100 0200 0300 0400 0500 0600 0700 0800 0900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000 2100 2200 2300 2400

20

19

18

17

16

15

14

13

12

11

10

9

8

7

6

5

4

3

2

1

0

EURUSD Time of Day Chart (Mondays)2-, 7- and 15-Minute Range (1/1/2005 to 4/14/2006)

Figure H-2 Monday composite range chart.

252 Appendices

0000 0100 0200 0300 0400 0500 0600 0700 0800 0900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000 2100 2200 2300 2400

EURUSD Time of Day Chart (Tuesdays)2-, 7- and 15-Minute Range (1/1/2005 to 4/14/2006) 26

252423222120191817161514131211109876543210

Figure H-3 Tuesday composite range chart.


EURUSD Time of Day Chart (Wednesdays)2-, 7- and 15-Minute Range (1/1/2005 to 4/14/2006)

2627

252423222120191817161514131211109876543210

0000 0100 0200 0300 0400 0500 0600 0700 0800 0900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000 2100 2200 2300 2400

Figure H-4 Wednesday composite range chart.

254 Appendices

EURUSD Time of Day Chart (Thursdays)2-, 7- and 15-Minute Range (1/1/2005 to 4/14/2006)

2627282930313233

25242322212019181716151413121110

9876543210

0000 0100 0200 0300 0400 0500 0600 0700 0800 0900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000 2100 2200 2300 2400

Figure H-5 Thursday composite range chart.


EURUSD Time of Day Chart (Fridays)2-, 7- and 15-Minute Range (1/1/2005 to 4/14/2006)

525456

50484644424038363432302826242220181614121086420

0000 0100 0200 0300 0400 0500 0600 0700 0800 0900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000 2100 2200 2300 2400

Figure H-6 Friday composite range chart.

Appendix IComposite

Activity Charts(Weekly)

This and the following two appendices are the composite of sixday-of-the-week charts (Sunday through Friday) using the same statistical method (activity, range, or momentum). Saturdays arenot included because the liquidity of trading diminishes drasticallyat 5:30 p.m. ET Friday.

The time frame again spans from January 1, 2005 through Aptil14, 2006. Time-of-day charts employed 1-minute raw data intervals,whereas day-of-week charts use hourly data as the basis.

The currency pairs in these three appendices are arranged inthe following order:

EURUSDGBPUSDUSDCHFUSDJPYUSDCADAUDUSDNZDUSDUSDSEKUSDNOKUSDDKK

257


258 Appendices

The vertical numeric scale on the right is activity expressed intotal number of ticks (upticks plus downticks) during each timeinterval. Each lower (dark) vertical bar represents the activity forthe current hour. The sum of the two vertical bars (light and dark)represents the activity for each 2-hour period (the current hourplus the previous hour).

EURUSD Day of Week Chart1- and 2-Hour Activity (1/1/2005 to 4/14/2006)

2300022000210002000019000180001700016000150001400013000120001100010000900080007000600050004000300020001000

FriThuWedTueMonSun

0

Figure I-1 EURUSD composite activity chart.

Composite Activity Charts (Weekly) 259

GBPUSD Day of Week Chart1- and 2-Hour Activity (1/1/2005 to 4/14/2006)

2600027000

2400023000

25000

22000210002000019000180001700016000150001400013000120001100010000900080007000600050004000300020001000

FriThuWedTueMonSun

0

Figure I-2 GBPUSD composite activity chart.

260 Appendices

USDCHF Day of Week Chart1- and 2-Hour Activity (1/1/2005 to 4/14/2006)

FriThuWedTueMonSun

240002300022000210002000019000180001700016000150001400013000120001100010000900080007000600050004000300020001000

0

Figure I-3 USDCHF composite activity chart.


USDJPY Day of Week Chart1- and 2-Hour Activity (1/1/2005 to 4/14/2006)

FriThuWedTueMonSun

19000

18000

17000

16000

15000

14000

13000

12000

11000

10000

9000

8000

7000

6000

5000

4000

3000

2000

1000

0

Figure I-4 USDJPY composite activity chart.

262 Appendices

USDCAD Day of Week Chart1- and 2-Hour Activity (1/1/2005 to 4/14/2006)

FriThuWedTueMonSun

16000

15000

14000

13000

12000

11000

10000

9000

8000

7000

6000

5000

4000

3000

2000

1000

0

Figure I-5 USDCAD composite activity chart.


AUDUSD Day of Week Chart1- and 2-Hour Activity (1/1/2005 to 4/14/2006)

FriThuWedTueMonSun

12000

11000

10000

9000

8000

7000

6000

5000

4000

3000

2000

1000

0

Figure I-6 AUDUSD composite activity chart.

264 Appendices

NZDUSD Day of Week Chart1- and 2-Hour Activity (1/1/2005 to 4/14/2006)

FriThuTue WedMonSun

8000

7000

6000

5000

4000

3000

2000

1000

0

Figure I-7 NZDUSD composite activity chart.


USDSEK Day of Week Chart1- and 2-Hour Activity (1/1/2005 to 4/14/2006)

FriThuWedTueMonSun

10000

9000

8000

7000

6000

5000

4000

3000

2000

1000

0

Figure I-8 USDSEK composite activity chart.

266 Appendices

USDNOK Day of Week Chart1- and 2-Hour Activity (1/1/2005 to 4/14/2006)

FriThuTue WedMonSun

10000

9000

8000

7000

6000

5000

4000

3000

2000

1000

0

Figure I-9 USDNOK composite activity chart.


USDDKK Day of Week Chart1- and 2-Hour Activity (1/1/2005 to 4/14/2006)

FriThuWedTueMonSun

8000

7000

6000

5000

4000

3000

2000

1000

0

Figure I-10 USDDKK composite activity chart.

Appendix JComposite Range

Charts (Weekly)

See the comments at the beginning of Appendix I for a generaldescription of weekly composite charts.

The vertical numeric scale on the right of the composite rangechart is the absolute range expressed in total number of pips dur-ing each time interval. Each lower (dark) vertical bar representsthe range for the current hour. The sum of the two vertical bars(light and dark) represents the range for each two-hour period(the current hour plus the previous hour).

Pips are expressed in units of the quote (rightmost) currencyin the pair and calculated by multiplying the exchange rate by thecorresponding pip conversion factor (see Table J-1).

269

TABLE J-1 Pip Conversion Factors

Pair Factor

EURUSD 10,000GBPUSD 10,000USDCHF 10,000USDJPY 100USDCAD 10,000AUDUSD 10,000NZDUSD 10,000USDSEK 1,000USDNOK 1,000USDDKK 1,000


270 Appendices


02468

10121416182022242628303234363840424446485052545658606264666870

EURUSD Day of Week Chart1- and 2- Hour Range (1/1/2005 to 4/14/2006)

Figure J-1 EURUSD composite range chart.

Composite Range Charts (Weekly) 271

FriThuTue WedMonSun

87848178757269666360575451484542393633302724211815129630

GBPUSD Day of Week Chart1- and 2-Hour Range (1/1/2005 to 4/14/2006)

Figure J-2 GBPUSD composite range chart.

272 Appendices

USDCHF Day of Week Chart1- and 2-Hour Range (1/1/2005 to 4/14/2006)

FriThuTue WedMonSun

0369

12151821242730333639424548515457606366697275

Figure J-3 USDCHF composite range chart.


USDJPY Day of Week Chart1- and 2-Hour Range (1/1/2005 to 4/14/2006)

FriThuWedTueMonSun

57

54

51

48

45

42

39

36

33

30

27

24

21

18

15

12

9

6

3

0

Figure J-4 USDJPY composite range chart.

274 Appendices

USDCAD Day of Week Chart1- and 2-Hour Range (1/1/2005 to 4/14/2006)

FriThuTue WedMonSun

66

63

60

57

54

51

48

45

42

39

36

33

30

27

24

21

18

15

12

9

6

3

0

Figure J-5 USDCAD composite range chart.


0

2

4

6

8

10

12

14

16

18

20

22

24

26

28

30

32

34

36

38

40

42


AUDUSD Day of Week Chart1- and 2- Hour Range (1/1/2005 to 4/14/2006)

Figure J-6 AUDUSD composite range chart.

276 Appendices

FriThuTue WedMonSun

42

40

38

36

34

32

30

28

26

24

22

20

18

16

14

12

10

8

6

4

2

0

NZDUSD Day of Week Chart1- and 2-Hour Range (1/1/2005 to 4/14/2006)

Figure J-7 NZDUSD composite range chart.


USDSEK Day of Week Chart1- and 2-Hour Range (1/1/2005 to 4/14/2006)

FriThuWedTueMonSun

5250484644424038363432302826242220181614121086420

Figure J-8 USDSEK composite range chart.

278 Appendices

USDNOK Day of Week Chart1- and 2-Hour Range (1/1/2005 to 4/14/2006)

FriThuTue WedMonSun

0

2

4

6

8

10

12

14

16

18

20

22

24

26

28

30

32

34

36

38

40

Figure J-9 USDNOK composite range chart.


FriThuTue WedMonSun

0

2

4

6

8

10

12

14

16

18

20

22

24

26

28

30

32

34

36

USDDKK Day of Week Chart1- and 2-Hour Range (1/1/2005 to 4/14/2006)

Figure J-10 USDDKK composite range chart.

Appendix KResources

PERIODICALS

Active Trader (TechInfo, Inc.), www.activetradermag.com.Currency Trader, www.currencytradermag.com.EForex, www.eforex.net.Euromoney, www.euromoney.com.Futures (Futures Magazine, Inc.), www.futuresmag.com.FX&MM, www.russellpublishing.com/FX&MM/index.html.FX Week, www.fxweek.com.Technical Analysis of Stocks & Commodities, www.traders.com.Traders Journal, www.traders-journal.com.

BOOKS

Aby, Carroll D, Jr., Point and Figure Charting. Greenville, SC: TradersPress, 1996.

Archer, Michael, & James Bickford, Getting Started in CurrencyTrading. Hoboken, NJ: Wiley, 2004.

Bickford, James, Chart Plotting Techniques for Technical Analysts.Boulder, CO: Syzygy, 2002.

Bigalow, Stephen, Profitable Candlestick Trading. New York: Wiley,2002.

Bulkowski, Thomas, Encyclopedia of Chart Patterns. Hoboken, NJ:Wiley, 2005.

281


www.activetradermag.com

www.currencytradermag.com

www.eforex.net

www.euromoney.com

www.futuresmag.com

www.russellpublishing.com/FX&MM/index.html

www.fxweek.com

www.traders.com

www.traders-journal.com

Dorsey, Thomas, Point & Figure Charting. New York: Wiley, 1995.Henderson, Callum, Currency Strategy. New York: Wiley, 2002.Horner, Raghee, Forex Trading for Maximum Profit. Hoboken, NJ:

Wiley, 2005.Klopfenstein, Gary, Trading Currency Cross Rates. New York: Wiley,

1993.Lien, Kathy, Day Trading the Currency Market. New York: Wiley, 2004.Lindsay, Charles, Trident: A Trading Strategy. Trident, 1976.Luca, Cornelius, Technical Analysis Applications in the Global Currency

Markets. Englewood Cliffs, NJ: Prentice-Hall, 2000.Luca, Cornelius, Trading in the Global Currency Markets. Englewood

Cliffs, NJ: Prentice-Hall, 2000.McGee, John, Technical Analysis of Stock Trends. New York, NY:

American Management Association, 2001.Murphy, John, Intermarket Financial Analysis. New York: Wiley, 2000.Murphy, John, Technical Analysis of the Financial Markets. Englewwod

Cliffs, NJ: Prentice-Hall, 1999.Nison, Steve, Japanese Candlestick Charting Techniques. New York, NY:

Prentice-Hall, 2001.Nofri, Eugene, & Jeanette Nofri-Steinberg, Success in Commodities.

Success, 1975.Reuters, Ltd., An Introduction to Foreign Exchange & Money Markets.

Reuters, 1999Rosenstreich, Peter, Forex Revolution. Englewood Cliffs, NJ:

Prentice-Hall, 2004.Ross, Joe, Trading by the Minute. Joe Ross, 1991.Schlossberg, Boris, Technical Analysis of the Currency Market.

Hoboken, NJ: Wiley, 2006.Shamah, Shani, A Foreign Exchange Primer. New York: Wiley, 2003.Zieg, Kermit, Point & Figure. Greenville, SC: Traders Press, 1997.The world’s largest supplier of mail-order investment/trading

books is Traders Press at www.traderspress.com.

282 Appendices

www.traderspress.com

Glossary of Forex Terms

A

Appreciation A currency is said to appreciate when it strengthensin price in response to market demand.

Arbitrage The purchase or sale of an instrument and simultane-ous taking of an equal and opposite position in a related market inorder to take advantage of small price differentials between markets.

Ask price The price at which the market is prepared to sell a spe-cific currency in a foreign exchange contract or cross-currency con-tract. At this price, the trader can buy the base currency. In thequotation, it is shown on the right side. For example, in the quoteUSDCHF 1.4527/32, the ask price is 1.4532, meaning that you canbuy one U.S. dollar for 1.4532 Swiss francs.

At best An instruction given to a dealer to buy or sell at the bestrate that can be obtained.

At or better An order to deal at a specific rate or better.

B

Balance of trade The value of a country’s exports minus itsimports.

Bar chart A type of chart that consists of four significant points:the high and low prices, which form the vertical bar; the opening

283


284 Glossary

price, which is marked with a little horizontal line to the left of thebar; and the closing price, which is marked with a little horizontalline to the right of the bar.

Base currency The first currency in a currency pair. It shows howmuch the base currency is worth, as measured against the secondcurrency. For example, if the USDCHF rate equals 1.6215, then oneUSD is worth CHF 1.6215. In the foreign exchange markets, theU.S. dollar is normally considered the “base” currency for quotes,meaning that quotes are expressed as a unit of 1 USD per the othercurrency quoted in the pair. The primary exceptions to this ruleare the British pound, the euro, and the Australian dollar.

Bear market A market distinguished by declining prices.

Bid price The bid is the price at which the market is preparedto buy a specific currency in a foreign exchange contract or cross-currency contract. At this price, the trader can sell the base currency. It is shown on the left side of the quotation. For exam-ple, in the quote USDCHF 1.4527/32, the bid price is 1.4527;meaning that you can sell 1 USD for 1.4527 Swiss francs.

Bid/ask spread The difference between the bid and offer price.

Big figure quote Dealer expression referring to the first few digitsof an exchange rate. These digits are often omitted in dealer quotes.For example, a USDJPY rate might be 117.30/117.35 but would bequoted verbally without the first three digits, that is, 30/35.

BLS Bureau of Labor Statistics.

Book In a professional trading environment, a book is the sum-mary of a trader’s or desk’s total positions.

Broker An individual or firm that acts as an intermediary, puttingtogether buyers and sellers for a fee or commission. In contrast, adealer commits capital and takes one side of a position, hoping toearn a spread (profit) by closing out the position in a subsequenttrade with another party.

Bretton Woods Agreement of 1944 An agreement that establishedfixed foreign exchange rates for major currencies, provided for cen-tral bank intervention in the currency markets, and pegged the price

of gold at US$35 per ounce. The agreement lasted until 1971, whenPresident Nixon overturned the Bretton Woods agreement andestablished a floating exchange rate for the major currencies.

Bull market A market distinguished by rising prices.

Bundesbank Germany’s Central Bank.

C

Cable Trader jargon referring to the sterling/USD exchangerate. So-called because the rate was transmitted originally via atrans-Atlantic cable beginning in the mid-1800s.

Candlestick chart A chart that indicates the trading range for theday as well as the opening and closing prices. If the open price ishigher than the close price, the rectangle between the open andclose price is shaded. If the close price is higher than the openprice, that area of the chart is not shaded.

Cash market The market in the actual financial instrument onwhich a futures or options contract is based.

Central bank A government or quasi-governmental organizationthat manages a country’s monetary policy. For example, the U.S.central bank is the Federal Reserve, and the German central bankis the Bundesbank.

CFTC Commodity Futures Trading Commission.

Chartist An individual who uses charts and graphs and interpretshistorical data to find trends and predict future movements. Alsoreferred to as technical trader.

Cleared funds Funds that are freely available, sent in to settle a trade.

Closed position Exposures in foreign currencies that no longerexist. The process to close a position is to sell or buy a certainamount of currency to offset an equal amount of the open posi-tion. This will square the position.

Clearing The process of settling a trade.

Glossary 285

CME Chicago Mercantile Exchange.

Contagion The tendency of an economic crisis to spread from onemarket to another. In 1997, political instability in Indonesia causedhigh volatility in their domestic currency, the rupiah. From there,the contagion spread to other Asian emerging currencies and thento Latin America and now is referred to as the Asian contagion.

Collateral Something given to secure a loan or as a guarantee of performance.

Commission A transaction fee charged by a broker.

Confirmation A document exchanged by counterparts to a trans-action that states the terms of said transaction.

Contract The standard unit of trading.

Counter currency The second listed currency in a currency pair.

Counterparty One of the participants in a financial transaction.

Country risk Risk associated with a cross-border transaction,including, but not limited to, legal and political conditions.

Cross-currency pair A foreign exchange transaction in which oneforeign currency is traded against a second foreign currency, forexample, EURGBP.

Cross-rate Same as cross-currency pair.

Currency Any form of money issued by a government or centralbank and used as legal tender and a basis for trade.

Currency pair The two currencies that make up a foreignexchange rate, for example, EURUSD.

Currency risk The probability of an adverse change in exchangerates.

D

Day trader Speculators who take positions in currencies that arethen liquidated prior to the close of the same trading day.

Dealer An individual or firm that acts as a principal or counter-part to a transaction. Principals take one side of a position, hoping

286 Glossary

to earn a spread (profit) by closing out the position in a subsequenttrade with another party. In contrast, a broker is an individual or firm that acts as an intermediary, putting together buyers andsellers for a fee or commission.

Deficit A negative balance of trade or payments.

Delivery A Forex trade where both sides make and take actualdelivery of the currencies traded.

Depreciation A fall in the value of a currency owing to marketforces.

Derivative A contract that changes in value in relation to theprice movements of a related or underlying security, future, orother physical instrument. An option is the most common deriv-ative instrument.

Devaluation The deliberate downward adjustment of a currency’sprice, normally by official announcement.

E

Economic indicator A government-issued statistic that indicatescurrent economic growth and stability. Common indicatorsinclude employment rates, gross domestic product (GDP), infla-tion, retail sales, and so forth.

ECU European Currency Unit; see EMU.

End-of-day order (EOD) An order to buy or sell at a specifiedprice. This order remains open until the end of the trading day,which is typically 5 p.m. ET.

European Monetary Union (EMU) One of th goals of the EMUwas to establish a single European currency called the euro, whichofficially replaced the national currencies of the member EU coun-tries in 2002. On Janaury 1, 1999, the transitional phase to intro-duce the euro began. The euro now exists as a banking currency,and paper financial transactions and foreign exchanges are madein euros. This transition period will last for three years, at whichtime euro notes and coins will enter circulation. On July 1, 2002,

Glossary 287

only euros will be legal tender for EMU participants, the national currencies of the member countries will cease to exist. The currentmembers of the EMU are Germany, France, Belgium, Luxembourg,Austria, Finland, Ireland, the Netherlands, Italy, Spain, andPortugal.

Euro The currency of the European Monetary Union (EMU). Areplacement for the European Currency Unit (ECU).

European Central Bank (ECB) The central bank for the new EMU.

F

Federal Deposit Insurance Corporation (FDIC) The regulatoryagency responsible for administering bank depository insurancein the United States.

Federal Reserve (Fed) The central bank for the United States.

First in first out (FIFO) Open positions are closed according tothe FIFO accounting rule. All positions opened within a particu-lar currency pair are liquidated in the order in which they wereoriginally opened.

Flat/square Dealer jargon used to describe a position that hasbeen completely reversed; for example, you bought $500,000 andthen sold $500,000, thereby creating a neutral (flat) position.

Foreign exchange (Forex, FX) The simultaneous buying of onecurrency and selling of another.

Forward The prespecified exchange rate for a foreign exchangecontract settling at some agreed future date based on the interest-rate differential between the two currencies involved.

Forward points The pips added to or subtracted from the cur-rent exchange rate to calculate a forward price.

Fundamental analysis Analysis of economic and political infor-mation with the objective of determining future movements in afinancial market.

Futures contract An obligation to exchange a good or instrumentat a set price on a future date. The primary difference between afuture and a forward is that futures typically are traded over an

288 Glossary

exchange (exchange-traded contacts, ETCs) versus forwards,which are considered over-the-counter (OTC) contracts. An OTCis any contract not traded on an exchange.

FX Foreign exchange.

G

G7 The seven leading industrial countries, namely, the UnitedStates, Germany, Japan, France, the United Kingdom, Canada, and Italy.

Going long The purchase of a stock, commodity, or currency forinvestment or speculation.

Going short The selling of a currency or instrument not ownedby the seller.

Good-’til-canceled order (GTC) An order to buy or sell at a spec-ified price. This order remains open until filled or until the clientcancels.

Gross domestic product (GDP) Total value of a country’s output,income, or expenditure produced within the country’s physicalborders.

Gross national product (GNP) Gross domestic product plusincome earned from investment or work abroad.

H

Hedge A position or combination of positions that reduces therisk of your primary position.

Hit the bid Acceptance of purchasing at the offer or selling atthe bid.

I

IMM International monetary market.

Inflation An economic condition whereby prices for consumergoods rise, eroding purchasing power.

Glossary 289

Initial margin The initial deposit of collateral required to enterinto a position as a guarantee on future performance.

Interbank rates The foreign exchange rates at which large inter-national banks quote other large international banks.

Intervention Action by a central bank to affect the value of its cur-rency by entering the market. Concerted intervention refers to actionby a number of central banks to control exchange rates.

K

Kiwi Slang for the New Zealand dollar.

L

Leading indicators Statistics that are considered to predict futureeconomic activity.

Leverage Also called margin. The ratio of the amount used in atransaction to the required security deposit.

LIBOR The London Inter-Bank Offered Rate. Banks use LIBORwhen borrowing from another bank.

Limit order An order with restrictions on the maximum price tobe paid or the minimum price to be received. As an example, ifthe current price of USDYEN is 117.00/05, then a limit order tobuy USD would be at a price below 102 (i.e., 116.50).

Liquidation The closing of an existing position through the exe-cution of an offsetting transaction.

Liquidity The ability of a market to accept large transaction withminimal to no impact on price stability; also, the ability to enterand exit a market quickly.

Long position A position that appreciates in value if marketprices increase. When the base currency in the pair is bought, theposition is said to be long.

Lot A unit to measure the amount of the deal. The value of thedeal always corresponds to an integer number of lots.

290 Glossary

M

Margin The required equity that an investor must deposit to collateralize a position.

Margin call A request from a broker or dealer for additionalfunds or other collateral to guarantee performance on a positionthat has moved against the customer.

Market maker A dealer who regularly quotes both bid and askprices and is ready to make a two-sided market for any financialinstrument.

Market risk Exposure to changes in market prices.

Mark-to-market Process of revaluating all open positions with thecurrent market prices. These new values then determine marginrequirements.

Maturity The date for settlement or expiry of a financial instru-ment.

N

Net position The amount of currency bought or sold that has notyet been offset by opposite transactions.

NFA National Futures Association.

O

Offer The rate at which a dealer is willing to sell a currency. Seeask price.

Offsetting transaction A trade that serves to cancel or offset someor all of the market risk of an open position.

One cancels the other order (OCO) A designation for two orderswhereby if one part of the two orders is executed, the other is auto-matically canceled.

Open order An order that will be executed when a market movesto its designated price. Normally associated with good-’til-can-celed orders.

Glossary 291

Open position An active trade with corresponding unrealizedprofit and loss that has not been offset by an equal and opposite deal.

Over the counter (OTC) Used to describe any transaction thatis not conducted over an exchange.

Overnight position A trade that remains open until the next busi-ness day.

Order An instruction to execute a trade at a specified rate.

P

Pip The smallest unit of price for any foreign currency. Digitsadded to or subtracted from the fourth decimal place, that is,0.0001. Also called point.

Political risk Exposure to changes in governmental policy thatwill have an adverse effect on an investor’s position.

Position The netted total holdings of a given currency.

Premium In the currency markets, describes the amount bywhich the forward or futures price exceeds the spot price.

Price transparency Describes quotes to which every market par-ticipant has equal access.

Profit /loss or P/L or gain/loss The actual realized gain or lossresulting from trading activities on closed positions, plus the the-oretical “unrealized” gain or loss on open positions that have beenmarked-to-market.

Q

Quote An indicative market price, normally used for informationpurposes only.

R

Rally A recovery in price after a period of decline.

Range The difference between the highest and lowest price of afuture recorded during a given trading session.

292 Glossary

Rate The price of one currency in terms of another, typically usedfor dealing purposes.

Resistance A term used in technical analysis indicating a specificprice level at which analysis concludes people will sell.

Revaluation An increase in the exchange rate for a currency asa result of central bank intervention. Opposite of devaluation.

Risk Exposure to uncertain change, most often used with a neg-ative connotation of adverse change.

Risk management The employment of financial analysis andtrading techniques to reduce and control exposure to various typesof risk.

Rollover Process whereby the settlement of a deal is rolled for-ward to another value date. The cost of this process is based onthe interest-rate differential of the two currencies.

Round trip Buying and selling of a specified amount of currency.

S

SEC Securities Exchange Commission.

Settlement The process by which a trade is entered into thebooks and records the counterparts to a transaction. The settle-ment of currency trades may or may not involve the actual physi-cal exchange of one currency for another.

Short position An investment position that benefits from adecline in market price. When the base currency in the pair is sold,the position is said to be short.

Spot price The current market price. Settlement of spot trans-actions usually occurs within two business days.

Spread The difference between the bid and offer prices.

Square Purchases and sales are in balance, and thus the dealerhas no open position.

Sterling Slang for British pound.

Stop-loss order Order type whereby an open position is auto-matically liquidated at a specific price. Often used to minimize

Glossary 293

exposure to losses if the market moves against an investor’s position. As an example, if an investor is long USD at 156.27, heor she might wish to put in a stop-loss order for 155.49, whichwould limit losses should the dollar depreciate, possibly below155.49.

Support levels A technique used in technical analysis that indi-cates a specific price ceiling and floor at which a given exchangerate will correct itself automatically. Opposite of resistance.

Swap A currency swap is the simultaneous sale and purchase ofthe same amount of a given currency at a forward exchange rate.

Swissy Market slang for Swiss franc.

T

Technical analysis An effort to forecast prices by analyzing mar-ket data, that is, historical price trends and averages, volumes, openinterest, and so forth.

Tick A minimum change in time required for the price tochange, up or down.

Transaction cost The cost of buying or selling a financial instru-ment.

Transaction date The date on which a trade occurs.

Turnover The total money value of all executed transactions ina given time period; volume.

Two-way price When both a bid rate and offer rate are quotedfor a Forex transaction.

U

Unrealized gain/loss The theoretical gain or loss on open posi-tions valued at current market rates, as determined by the brokerin its sole discretion. Unrealized gains/losses become profits/losseswhen position is closed.

294 Glossary

Uptick A new price quote at a price higher than the precedingquote.

Uptick rule In the United States, a regulation whereby a securitymay not be sold short unless the last trade prior to the short sale was at a price lower than the price at which the short sale isexecuted.

U.S. prime rate The interest rate at which U.S. banks will lendto their prime corporate customers.

V

Value date The date on which counterparts to a financial trans-action agree to settle their respective obligations, that is, exchang-ing payments. For spot currency transactions, the value date isnormally two business days forward. Also known as maturity date.

Variation margin Funds a broker must request from the client tohave the required margin deposited. The term usually refers toadditional funds that must be deposited as a result of unfavorableprice movements.

Volatility A statistical measure of a market’s price movementsover time characterized by deviations from a predetermined cen-tral value (usually the arithmetic mean).

W

Whipsaw Slang for a condition where any securities marketbegins moving laterally exhibiting very little volatility.

Y

Yard Slang for a billion.

Glossary 295

Absolute range, 55Activity, 43–50

author’s commentary, 50comparative, 45–48composite activity charts,

51–53, 241–247, 257–267daily composite activity

charts, 241–247daily OHLC and activity

charts, 229–239defined, 44direction, 48–50equiactivity chart, 44–45monthly OHLC and activity

charts, 217–228overview, 43range, and, 63–66single-currency, 48weekly composite activity

charts, 257–267Advanced features, 125–130

channel lines, 127–128cycle collapsing, 128–130fractal levels, 125–126pivot chart, 126–127

ADX, 49Aggressor, 34ANZAC Day, 215Arms, Richard W., Jr., 44–45AUDUSD composite activity

chart, 263

AUDUSD composite rangechart, 275

AUDUSD daily OHLC andactivity, 235

AUDUSD monthly OHLC andactivity, 224

Author (Bickford), 297Average directional index

(ADX), 49Average interval range, 61–62Average range, 150Average range interval

chart, 62

Balance of trade (BOT), 11–16author’s commentary,

12–14, 16defined, 11descriptive statistics, 14differential analysis, 15enhancements, 15forecasting model, 11–12fundamental data, 12interpretation, 15regression results, 14–15serial correlation, 16smoothing techniques, 16triangular correlations, 16visual representation, 12

Bank holidays, 212–215Banking hours, 211–215

297

Index


298 Index

Basic three-wave cycles, 91–97bear cycle conversion, 96, 97connector, 95, 96contracting ascending

triangle, 93contracting descending

triangle, 93contracting symmetrical

triangle, 93–94expanding ascending

triangle, 94expanding descending

triangle, 94, 95expanding symmetrical

triangle, 95frequencies, 96–97impulse cycle, 92overview, 91rectangle, 92

Bear cycle, 91Bear cycle conversion, 96, 97Bickford, James L., 297Blind trading, 195–197Books, 281–282BOT, 11–16. See also Balance

of trade (BOT)Boxing Day, 215Bull cycle, 91Bull impulse, 130Bull rectangle, 130

Channel lines, 127–128Channeling, 127Chimera cycle, 123–124Chimera cycle forecast, 123Collapsing bull cycles, 130Combined inertia, 72

Comparative activity, 45–48Composite activity charts,

51–53, 241–247, 257–267Composite modeling, 25–26Composite range charts,

59–62, 249–255, 269–279Connector, 95, 96, 129Consumer price index

(CPI), 17–24CPI data, 18cross-correlation, 22–23currency data, 21enhancements, 23–24forecasting model, 18independent variable, 21individual linear

regressions, 18overview, 17regression results, 20

Contracting ascending triangle, 93, 129, 130

Contracting descending triangle, 93, 129

Contracting symmetrical triangle, 93–94, 129

CPI, 17–24. See also Consumerprice index (CPI)

Cycle, 85Cycle collapsing, 128–130Cycle preliminaries,

85–89definitions, 85minimum fluctuation

unit, 86minimum reversal

amount, 87swing reversal

algorithm, 87–89

Daily composite activity charts,241–247

Daily composite range charts,249–255

Daily OHLC and activitycharts, 229–239

Day-of-week activity chart, 52, 53Day-of-week range chart, 61Definitions (glossary), 283–295Direction, 48–50Direction oscillator chart, 49Disk Trading, Ltd. (DTL), 43Domestic economic

calendar, 38Dominance/dependence

relationship, 16Dow, Charles, 82DTL, 43

EBS, 31–32Economic indicators, 37–39Economic news announce-

ments, 151ECU, 210Electronic broking, 31–36

EBS, 31–32live book, 34–36market makers, 33order flow, 33–34Reuters, 32

Electronic Broking Services(EBS), 31–32

Elliott, Ralph N., 77, 79–82, 85Elliott corrective pattern, 80Elliott cycle names, 82Elliott impulse pattern, 80Elliott wave, 79–83Equiactivity chart, 44–45

Euro, 210Euro currency, 209–210Euro currency and

fundamentals, 26–27European Currency Unit

(ECU), 210EURUSD composite activity

chart, 258EURUSD composite range

chart, 270EURUSD daily OHLC and

activity, 230EURUSD monthly OHLC and

activity, 219Exchange rates, 207–208Expanding ascending triangle,

94, 129, 130Expanding descending

triangle, 94, 95, 129Expanding symmetrical

triangle, 95, 129Extended contracting symmet-

rical triangle, 115Extended contracting triangle,

115–116Extended expanding symmet-

rical triangle, 116Extended expanding triangle,

116–117Extended impulse cycle,

113–114, 120–122

Fifteen-pip minimum reversalamount, 193

Fifth wave, 111–117. See alsoForecasting the fifth wave

Forecasting the fifth wave,111–117

Index 299

Forecasting …(contd.)extended contracting

triangle, 115–116extended expanding

triangle, 116–117extended impulse cycle,

113–114fifth-wave forecast, 113frequencies, 112potential head and

shoulders, 114–115pruning, 111–112

Forecasting the fourth wave, 107–110

Forecasting the sixth wave, 119–124chimera cycle, 123–124extended impulse cycle,

120–122frequencies, 119–120head and shoulders,

122–123overview, 119

Forecasting the third wave, 99–101

Foreign economic calendar, 38Forex inertia, 68–70Forex Wave Theory (Bickford), 77Four-wave bear cycle, 106Four-wave cycle

components, 104Fourth wave, 107–110Fractal geometry, 82Fractal levels, 125–126,

192–193Friday composite activity

chart, 247Friday composite range

chart, 255

Fundamental analysis, 4–5GBPUSD and regression

line, 9GBPUSD composite activity

chart, 259GBPUSD composite range

chart, 271GBPUSD currency pair, 8, 22GBPUSD daily OHLC and

activity, 231GBPUSD monthly OHLC and

activity, 220Global banking hours,

211–215Glossary, 283–295

Head and shoulders, 114,122–123

Head and shoulders withdescending neckline, 114

ICAP, 31Impulse cycle, 92, 129Inertia, 67–73

author’s commentary, 72charts, 70–72Forex, 68–70momentum, and, 72–73overview, 67raw currency data, 67–68

Inertia charts, 70–72Informed trading, 197–198Intercapital Public Limited

Company (ICAP), 31Interdealer broker, 32Interest rate regression coeffi-

cients, 10Interest rates, 7–10

author’s commentary, 10

300 Index

forecasting model, 8fundamental data, 8regression results, 10visual representation, 8–9

International bank holidays,213–214

International foreignexchange rates, 207–208

ISO currency pairs, 201–205

Lateral congestion, 92Likelihood relationship, 99Live book, 34–36Logical length, 99

Market makers, 33Maxima and minima, 193–194Minimum fluctuation unit, 86Minimum reversal amount, 87Momentum, 67, 72–73Monday composite activity

chart, 243Monday composite range

chart, 251Monthly OHLC and activity

charts, 217–228Multiple-wave cycles, 103–106

New Concepts in TechnicalTrading Systems (Wilder), 49

News sources, 37NZDUSD composite activity

chart, 264NZDUSD composite range

chart, 276NZDUSD daily OHLC and

activity, 236NZDUSD monthly OHLC and

activity, 225

OHLC bar charts, 133–151author’s commentary,

151–152conventional chart, 46daily OHLC and activity

charts, 229–239monthly OHLC and activity

charts, 217–228overview, 133–134range ratio, 150–151sample charts, 134–148summary of statistics, 149

Order execution mechanism,33

Order flow, 33–34

Partial regression coefficientfor a multiple regression, 26

Peak, 85, 86Periodicals, 281Pivot chart, 126–127Potential head and shoulders,

114–115Preshockwave calm, 133Primary cycle types, 112Primary cycles, 111

Range, 55–66absolute, 55activity, and, 63–66author’s commentary, 58average interval, 61–62composite range charts,

59–62, 249–255, 269–279daily composite range

charts, 249–255relative, 56, 57weekly composite range

charts, 269–279

Index 301

Range ratio, 150–151Range ratio index, 150Range ratio oscillator, 151Reactive response, 133Rectangle, 92, 129Recurring wave patterns, 80–81Relative range, 56, 57Resources, 281–282Retail price index, 17. See also

Consumer price index (CPI)Retracement trading, 197Reuters, 32Reuters Dealing 3000 system, 32Reversal charts, 78–79

Scheduled news releases,37–39

Secondary cycles, 111Shockwave observations,

191–198blind trading, 195–197caveat, 198fractal levels, 192–193informed trading, 197–198maxima and minima,

193–194overview, 191retracement trading, 197

Shockwave surge, 133Single-currency activity, 48Six-wave cycle, 105Sixth wave, 119–124. See also

Forecasting the sixth wave

Solution to simultaneous equations, 26

Streaming spot currency data, 43

Sunday composite activitychart, 242

Sunday composite range chart, 250

Swing charts, 173–189Swing reversal algorithm,

87–89

Technical analysis, 4Ten-pip minimum reversal

amount, 192Terminology (glossary),

283–295Third wave, 99–101Three-wave bull connector

cycle, 109Three-wave cycle

frequencies, 97Three-wave cycles, 91–97. See

also Basic three–wave cyclesThursday composite activity

chart, 246Thursday composite range

chart, 254Tick charts, 153–170

author’s commentaries,169–170

overview, 153sample charts, 154–168summary of statistics, 169

Time frame, 150Time interval, 150Time-of-day activity chart, 51, 52Time-of-day range chart, 59, 60Total range, 150Trough, 85Tuesday composite activity

chart, 244

302 Index

Tuesday composite rangechart, 252

U.K. consumer price index,19, 20

U.S.-Canada balance of trade, 13

U.S. consumer price index, 19USDCAD composite activity

chart, 262USDCAD composite range

chart, 274USDCAD currency pair, 11USDCAD daily OHLC and

activity, 234USDCAD monthly midrange, 13USDCAD monthly OHLC and

activity, 223USDCHF composite activity

chart, 260USDCHF composite range

chart, 272USDCHF daily OHLC and

activity, 232USDCHF monthly OHLC and

activity, 221USDDKK composite activity

chart, 267USDDKK composite range

chart, 279USDDKK daily OHLC and

activity, 239USDDKK monthly OHLC and

activity, 228USDJPY composite activity

chart, 261USDJPY composite range

chart, 273

USDJPY daily OHLC and activity, 233

USDJPY monthly OHLC andactivity, 222

USDNOK composite activitychart, 266

USDNOK composite rangechart, 278

USDNOK daily OHLC andactivity, 238

USDNOK monthly OHLC andactivity, 227

USDSEK composite activitychart, 265

USDSEK composite rangechart, 277

USDSEK daily OHLC andactivity, 237

USDSEK monthly OHLC andactivity, 226

Valley, 85, 86Volume Cycles in the Stock Market

(Arms), 45

Wave, 85Wave principles, 77–83

Elliott wave, 79–83overview, 77recurring wave patterns,

80–81reversal charts, 78–79waves within waves, 81–82

Waves within waves, 81–82Wednesday composite activity

chart, 245Wednesday composite range

chart, 253

Index 303

About the Author

James L. Bickford began trading commodity futures in the early1980s. Since 2001, he has focused his investment efforts solely ontrading spot currency pairs via online Forex dealers. His academicbackground is in computer science, applied mathematics, and sta-tistics. By profession, he has been a software engineer for nearlythree decades. He has authored (or coauthored) several books ontechnical analysis and currency trading: Chart Plotting Algorithms forTechnical Analysts (2001), Getting Started in Currency Trading (2005),Forex Chartist’s Companion: A Visual Approach to Technical Analysis(2006), Forex Trader’s Companion: Focus on Major Currencies (2007),and Forex Wave Theory (2007).

Weekly composite activitycharts, 257–267

Weekly composite rangecharts, 269–279

Whit Monday, 215

Wilder, J. Welles, 49World currencies, 201–205

ZGZ6, 34

304 Index


FOREX - cabafx.comcabafx.com/trading-ebooks-collection/JAMES L. BICKFORD - Forex... · FOREX...

Documents

Transcript of FOREX - cabafx.comcabafx.com/trading-ebooks-collection/JAMES L. BICKFORD - Forex... · FOREX...